Full Publication List hover over images to zoom-in —
auto-generated by
2024
(6)
|
Cone-Traced Supersampling With Subpixel Edge Reconstruction.
Chubarau, A.; Zhao, Y.; Rao, R.; Nowrouzezahrai, D.; and Kry, P. G.
IEEE Trans. Vis. Comput. Graph., 30(9): 6421–6432. 2024.
Paper
doi
link
bibtex
@article{DBLP:journals/tvcg/ChubarauZRNK24,
author = {Andrei Chubarau and
Yangyang Zhao and
Ruby Rao and
Derek Nowrouzezahrai and
Paul G. Kry},
title = {Cone-Traced Supersampling With Subpixel Edge Reconstruction},
journal = {{IEEE} Trans. Vis. Comput. Graph.},
volume = {30},
number = {9},
pages = {6421--6432},
year = {2024},
url = {https://doi.org/10.1109/TVCG.2023.3343166},
doi = {10.1109/TVCG.2023.3343166},
timestamp = {Thu, 22 Aug 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tvcg/ChubarauZRNK24.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Minimax Exploiter: A Data Efficient Approach for Competitive Self-Play.
Bairamian, D.; Marcotte, P.; Romoff, J.; Robert, G.; and Nowrouzezahrai, D.
In Dastani, M.; Sichman, J. S.; Alechina, N.; and Dignum, V., editor(s), Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2024, Auckland, New Zealand, May 6-10, 2024, pages 114–122, 2024. International Foundation for Autonomous Agents and Multiagent Systems / ACM
Paper
doi
link
bibtex
2 downloads
@inproceedings{DBLP:conf/atal/BairamianMRRN24,
author = {Daniel Bairamian and
Philippe Marcotte and
Joshua Romoff and
Gabriel Robert and
Derek Nowrouzezahrai},
editor = {Mehdi Dastani and
Jaime Sim{\~{a}}o Sichman and
Natasha Alechina and
Virginia Dignum},
title = {Minimax Exploiter: {A} Data Efficient Approach for Competitive Self-Play},
booktitle = {Proceedings of the 23rd International Conference on Autonomous Agents
and Multiagent Systems, {AAMAS} 2024, Auckland, New Zealand, May 6-10,
2024},
pages = {114--122},
publisher = {International Foundation for Autonomous Agents and Multiagent Systems
/ {ACM}},
year = {2024},
url = {https://dl.acm.org/doi/10.5555/3635637.3662858},
doi = {10.5555/3635637.3662858},
timestamp = {Wed, 26 Jun 2024 14:06:50 +0200},
biburl = {https://dblp.org/rec/conf/atal/BairamianMRRN24.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Model-Based Solution to the Offline Multi-Agent Reinforcement Learning Coordination Problem.
Barde, P.; Foerster, J.; Nowrouzezahrai, D.; and Zhang, A.
In Dastani, M.; Sichman, J. S.; Alechina, N.; and Dignum, V., editor(s), Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2024, Auckland, New Zealand, May 6-10, 2024, pages 141–150, 2024. International Foundation for Autonomous Agents and Multiagent Systems / ACM
Paper
doi
link
bibtex
3 downloads
@inproceedings{DBLP:conf/atal/BardeFNZ24,
author = {Paul Barde and
Jakob Foerster and
Derek Nowrouzezahrai and
Amy Zhang},
editor = {Mehdi Dastani and
Jaime Sim{\~{a}}o Sichman and
Natasha Alechina and
Virginia Dignum},
title = {A Model-Based Solution to the Offline Multi-Agent Reinforcement Learning
Coordination Problem},
booktitle = {Proceedings of the 23rd International Conference on Autonomous Agents
and Multiagent Systems, {AAMAS} 2024, Auckland, New Zealand, May 6-10,
2024},
pages = {141--150},
publisher = {International Foundation for Autonomous Agents and Multiagent Systems
/ {ACM}},
year = {2024},
url = {https://dl.acm.org/doi/10.5555/3635637.3662861},
doi = {10.5555/3635637.3662861},
timestamp = {Fri, 03 May 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/atal/BardeFNZ24.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Regional Adaptive Metropolis Light Transport.
Otsu, H.; Herveau, K.; Hanika, J.; Nowrouzezahrai, D.; and Dachsbacher, C.
CoRR, abs/2402.08273. 2024.
Paper
doi
link
bibtex
3 downloads
@article{DBLP:journals/corr/abs-2402-08273,
author = {Hisanari Otsu and
Killian Herveau and
Johannes Hanika and
Derek Nowrouzezahrai and
Carsten Dachsbacher},
title = {Regional Adaptive Metropolis Light Transport},
journal = {CoRR},
volume = {abs/2402.08273},
year = {2024},
url = {https://doi.org/10.48550/arXiv.2402.08273},
doi = {10.48550/ARXIV.2402.08273},
eprinttype = {arXiv},
eprint = {2402.08273},
timestamp = {Mon, 19 Feb 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2402-08273.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learnable Fractal Flames.
Bannister, J. J.; and Nowrouzezahrai, D.
CoRR, abs/2406.09328. 2024.
Paper
doi
link
bibtex
1 download
@article{DBLP:journals/corr/abs-2406-09328,
author = {Jordan J. Bannister and
Derek Nowrouzezahrai},
title = {Learnable Fractal Flames},
journal = {CoRR},
volume = {abs/2406.09328},
year = {2024},
url = {https://doi.org/10.48550/arXiv.2406.09328},
doi = {10.48550/ARXIV.2406.09328},
eprinttype = {arXiv},
eprint = {2406.09328},
timestamp = {Tue, 09 Jul 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2406-09328.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Reinforcement Learning for Sequence Design Leveraging Protein Language Models.
Subramanian, J.; Sujit, S.; Irtisam, N.; Sain, U.; Nowrouzezahrai, D.; Kahou, S. E.; and Islam, R.
CoRR, abs/2407.03154. 2024.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2407-03154,
author = {Jithendaraa Subramanian and
Shivakanth Sujit and
Niloy Irtisam and
Umong Sain and
Derek Nowrouzezahrai and
Samira Ebrahimi Kahou and
Riashat Islam},
title = {Reinforcement Learning for Sequence Design Leveraging Protein Language
Models},
journal = {CoRR},
volume = {abs/2407.03154},
year = {2024},
url = {https://doi.org/10.48550/arXiv.2407.03154},
doi = {10.48550/ARXIV.2407.03154},
eprinttype = {arXiv},
eprint = {2407.03154},
timestamp = {Wed, 07 Aug 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2407-03154.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2023
(13)
|
Learning neural implicit representations with surface signal parameterizations.
Guan, Y.; Chubarau, A.; Rao, R.; and Nowrouzezahrai, D.
Comput. Graph., 114: 257–264. 2023.
Paper
doi
link
bibtex
59 downloads
@article{DBLP:journals/cg/GuanCRN23,
author = {Yanran Guan and
Andrei Chubarau and
Ruby Rao and
Derek Nowrouzezahrai},
title = {Learning neural implicit representations with surface signal parameterizations},
journal = {Comput. Graph.},
volume = {114},
pages = {257--264},
year = {2023},
url = {https://doi.org/10.1016/j.cag.2023.06.013},
doi = {10.1016/J.CAG.2023.06.013},
timestamp = {Sun, 24 Sep 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cg/GuanCRN23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Differentiable visual computing for inverse problems and machine learning.
Spielberg, A.; Zhong, F.; Rematas, K.; Jatavallabhula, K. M.; Öztireli, C.; Li, T.; and Nowrouzezahrai, D.
Nat. Mac. Intell., 5(11): 1189–1199. 2023.
Paper
doi
link
bibtex
2 downloads
@article{DBLP:journals/natmi/SpielbergZRJOLN23,
author = {Andrew Spielberg and
Fangcheng Zhong and
Konstantinos Rematas and
Krishna Murthy Jatavallabhula and
Cengiz {\"{O}}ztireli and
Tzu{-}Mao Li and
Derek Nowrouzezahrai},
title = {Differentiable visual computing for inverse problems and machine learning},
journal = {Nat. Mac. Intell.},
volume = {5},
number = {11},
pages = {1189--1199},
year = {2023},
url = {https://doi.org/10.1038/s42256-023-00743-0},
doi = {10.1038/S42256-023-00743-0},
timestamp = {Sat, 13 Jan 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/natmi/SpielbergZRJOLN23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Visual question answering from another perspective: CLEVR mental rotation tests.
Beckham, C.; Weiss, M.; Golemo, F.; Honari, S.; Nowrouzezahrai, D.; and Pal, C.
Pattern Recognit., 136: 109209. 2023.
Paper
doi
link
bibtex
@article{DBLP:journals/pr/BeckhamWGHNP23,
author = {Christopher Beckham and
Martin Weiss and
Florian Golemo and
Sina Honari and
Derek Nowrouzezahrai and
Christopher Pal},
title = {Visual question answering from another perspective: {CLEVR} mental
rotation tests},
journal = {Pattern Recognit.},
volume = {136},
pages = {109209},
year = {2023},
url = {https://doi.org/10.1016/j.patcog.2022.109209},
doi = {10.1016/J.PATCOG.2022.109209},
timestamp = {Tue, 18 Apr 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/pr/BeckhamWGHNP23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
MeshDiffusion: Score-based Generative 3D Mesh Modeling.
Liu, Z.; Feng, Y.; Black, M. J.; Nowrouzezahrai, D.; Paull, L.; and Liu, W.
In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023, 2023. OpenReview.net
Paper
link
bibtex
1 download
@inproceedings{DBLP:conf/iclr/LiuFBNPL23,
author = {Zhen Liu and
Yao Feng and
Michael J. Black and
Derek Nowrouzezahrai and
Liam Paull and
Weiyang Liu},
title = {MeshDiffusion: Score-based Generative 3D Mesh Modeling},
booktitle = {The Eleventh International Conference on Learning Representations,
{ICLR} 2023, Kigali, Rwanda, May 1-5, 2023},
publisher = {OpenReview.net},
year = {2023},
url = {https://openreview.net/forum?id=0cpM2ApF9p6},
timestamp = {Tue, 06 Aug 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/iclr/LiuFBNPL23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Parameter-space ReSTIR for Differentiable and Inverse Rendering.
Chang, W.; Sivaram, V.; Nowrouzezahrai, D.; Hachisuka, T.; Ramamoorthi, R.; and Li, T.
In Brunvand, E.; Sheffer, A.; and Wimmer, M., editor(s), ACM SIGGRAPH 2023 Conference Proceedings, SIGGRAPH 2023, Los Angeles, CA, USA, August 6-10, 2023, pages 18:1–18:10, 2023. ACM
Paper
doi
link
bibtex
3 downloads
@inproceedings{DBLP:conf/siggraph/ChangSNHRL23,
author = {Wesley Chang and
Venkataram Sivaram and
Derek Nowrouzezahrai and
Toshiya Hachisuka and
Ravi Ramamoorthi and
Tzu{-}Mao Li},
editor = {Erik Brunvand and
Alla Sheffer and
Michael Wimmer},
title = {Parameter-space ReSTIR for Differentiable and Inverse Rendering},
booktitle = {{ACM} {SIGGRAPH} 2023 Conference Proceedings, {SIGGRAPH} 2023, Los
Angeles, CA, USA, August 6-10, 2023},
pages = {18:1--18:10},
publisher = {{ACM}},
year = {2023},
url = {https://doi.org/10.1145/3588432.3591512},
doi = {10.1145/3588432.3591512},
timestamp = {Sat, 05 Aug 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/siggraph/ChangSNHRL23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Efficient Graphics Representation with Differentiable Indirection.
Datta, S.; Marshall, C. S.; Dong, Z.; Li, Z.; and Nowrouzezahrai, D.
In Kim, J.; Lin, M. C.; and Bickel, B., editor(s), SIGGRAPH Asia 2023 Conference Papers, SA 2023, Sydney, NSW, Australia, December 12-15, 2023, pages 104:1–104:10, 2023. ACM
Paper
doi
link
bibtex
2 downloads
@inproceedings{DBLP:conf/siggrapha/DattaM0LN23,
author = {Sayantan Datta and
Carl S. Marshall and
Zhao Dong and
Zhengqin Li and
Derek Nowrouzezahrai},
editor = {June Kim and
Ming C. Lin and
Bernd Bickel},
title = {Efficient Graphics Representation with Differentiable Indirection},
booktitle = {{SIGGRAPH} Asia 2023 Conference Papers, {SA} 2023, Sydney, NSW, Australia,
December 12-15, 2023},
pages = {104:1--104:10},
publisher = {{ACM}},
year = {2023},
url = {https://doi.org/10.1145/3610548.3618203},
doi = {10.1145/3610548.3618203},
timestamp = {Thu, 01 Feb 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/siggrapha/DattaM0LN23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Neural Shadow Mapping.
Datta, S.; Nowrouzezahrai, D.; Schied, C.; and Dong, Z.
CoRR, abs/2301.05262. 2023.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2301-05262,
author = {Sayantan Datta and
Derek Nowrouzezahrai and
Christoph Schied and
Zhao Dong},
title = {Neural Shadow Mapping},
journal = {CoRR},
volume = {abs/2301.05262},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2301.05262},
doi = {10.48550/ARXIV.2301.05262},
eprinttype = {arXiv},
eprint = {2301.05262},
timestamp = {Thu, 13 Jul 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2301-05262.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
MeshDiffusion: Score-based Generative 3D Mesh Modeling.
Liu, Z.; Feng, Y.; Black, M. J.; Nowrouzezahrai, D.; Paull, L.; and Liu, W.
CoRR, abs/2303.08133. 2023.
Paper
doi
link
bibtex
1 download
@article{DBLP:journals/corr/abs-2303-08133,
author = {Zhen Liu and
Yao Feng and
Michael J. Black and
Derek Nowrouzezahrai and
Liam Paull and
Weiyang Liu},
title = {MeshDiffusion: Score-based Generative 3D Mesh Modeling},
journal = {CoRR},
volume = {abs/2303.08133},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2303.08133},
doi = {10.48550/ARXIV.2303.08133},
eprinttype = {arXiv},
eprint = {2303.08133},
timestamp = {Thu, 08 Aug 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2303-08133.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Model-Based Solution to the Offline Multi-Agent Reinforcement Learning Coordination Problem.
Barde, P.; Foerster, J.; Nowrouzezahrai, D.; and Zhang, A.
CoRR, abs/2305.17198. 2023.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2305-17198,
author = {Paul Barde and
Jakob Foerster and
Derek Nowrouzezahrai and
Amy Zhang},
title = {A Model-Based Solution to the Offline Multi-Agent Reinforcement Learning
Coordination Problem},
journal = {CoRR},
volume = {abs/2305.17198},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2305.17198},
doi = {10.48550/ARXIV.2305.17198},
eprinttype = {arXiv},
eprint = {2305.17198},
timestamp = {Wed, 07 Jun 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2305-17198.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Efficient Graphics Representation with Differentiable Indirection.
Datta, S.; Marshall, C. S.; Dong, Z.; Li, Z.; and Nowrouzezahrai, D.
CoRR, abs/2309.08387. 2023.
Paper
doi
link
bibtex
2 downloads
@article{DBLP:journals/corr/abs-2309-08387,
author = {Sayantan Datta and
Carl S. Marshall and
Zhao Dong and
Zhengqin Li and
Derek Nowrouzezahrai},
title = {Efficient Graphics Representation with Differentiable Indirection},
journal = {CoRR},
volume = {abs/2309.08387},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2309.08387},
doi = {10.48550/ARXIV.2309.08387},
eprinttype = {arXiv},
eprint = {2309.08387},
timestamp = {Thu, 01 Feb 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2309-08387.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
STAMP: Differentiable Task and Motion Planning via Stein Variational Gradient Descent.
Lee, Y.; Huang, P.; Jatavallabhula, K. M.; Li, A. Z.; Damken, F.; Heiden, E.; Smith, K.; Nowrouzezahrai, D.; Ramos, F.; and Shkurti, F.
CoRR, abs/2310.01775. 2023.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2310-01775,
author = {Yewon Lee and
Philip Huang and
Krishna Murthy Jatavallabhula and
Andrew Zou Li and
Fabian Damken and
Eric Heiden and
Kevin Smith and
Derek Nowrouzezahrai and
Fabio Ramos and
Florian Shkurti},
title = {{STAMP:} Differentiable Task and Motion Planning via Stein Variational
Gradient Descent},
journal = {CoRR},
volume = {abs/2310.01775},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2310.01775},
doi = {10.48550/ARXIV.2310.01775},
eprinttype = {arXiv},
eprint = {2310.01775},
timestamp = {Thu, 19 Oct 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2310-01775.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Minimax Exploiter: A Data Efficient Approach for Competitive Self-Play.
Bairamian, D.; Marcotte, P.; Romoff, J.; Robert, G.; and Nowrouzezahrai, D.
CoRR, abs/2311.17190. 2023.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2311-17190,
author = {Daniel Bairamian and
Philippe Marcotte and
Joshua Romoff and
Gabriel Robert and
Derek Nowrouzezahrai},
title = {Minimax Exploiter: {A} Data Efficient Approach for Competitive Self-Play},
journal = {CoRR},
volume = {abs/2311.17190},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2311.17190},
doi = {10.48550/ARXIV.2311.17190},
eprinttype = {arXiv},
eprint = {2311.17190},
timestamp = {Tue, 05 Dec 2023 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2311-17190.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Differentiable Visual Computing for Inverse Problems and Machine Learning.
Spielberg, A.; Zhong, F.; Rematas, K.; Jatavallabhula, K. M.; Öztireli, C.; Li, T.; and Nowrouzezahrai, D.
CoRR, abs/2312.04574. 2023.
Paper
doi
link
bibtex
2 downloads
@article{DBLP:journals/corr/abs-2312-04574,
author = {Andrew Spielberg and
Fangcheng Zhong and
Konstantinos Rematas and
Krishna Murthy Jatavallabhula and
Cengiz {\"{O}}ztireli and
Tzu{-}Mao Li and
Derek Nowrouzezahrai},
title = {Differentiable Visual Computing for Inverse Problems and Machine Learning},
journal = {CoRR},
volume = {abs/2312.04574},
year = {2023},
url = {https://doi.org/10.48550/arXiv.2312.04574},
doi = {10.48550/ARXIV.2312.04574},
eprinttype = {arXiv},
eprint = {2312.04574},
timestamp = {Wed, 03 Jan 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2312-04574.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2022
(19)
|
Dynamic Diffuse Global Illumination Resampling.
Majercik, Z.; Müller, T.; Keller, A.; Nowrouzezahrai, D.; and McGuire, M.
Comput. Graph. Forum, 41(1): 158–171. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/cgf/MajercikMKNM22,
author = {Zander Majercik and
Thomas M{\"{u}}ller and
Alexander Keller and
Derek Nowrouzezahrai and
Morgan McGuire},
title = {Dynamic Diffuse Global Illumination Resampling},
journal = {Comput. Graph. Forum},
volume = {41},
number = {1},
pages = {158--171},
year = {2022},
url = {https://doi.org/10.1111/cgf.14427},
doi = {10.1111/CGF.14427},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/cgf/MajercikMKNM22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Single-pass stratified importance resampling.
Ciklabakkal, E.; Gruson, A.; Georgiev, I.; Nowrouzezahrai, D.; and Hachisuka, T.
Comput. Graph. Forum, 41(4): 41–49. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/cgf/CiklabakkalGGNH22,
author = {Ege Ciklabakkal and
Adrien Gruson and
Iliyan Georgiev and
Derek Nowrouzezahrai and
Toshiya Hachisuka},
title = {Single-pass stratified importance resampling},
journal = {Comput. Graph. Forum},
volume = {41},
number = {4},
pages = {41--49},
year = {2022},
url = {https://doi.org/10.1111/cgf.14585},
doi = {10.1111/CGF.14585},
timestamp = {Mon, 08 Aug 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/CiklabakkalGGNH22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Efficient Differentiation of Pixel Reconstruction Filters for Path-Space Differentiable Rendering.
Yu, Z.; Zhang, C.; Nowrouzezahrai, D.; Dong, Z.; and Zhao, S.
ACM Trans. Graph., 41(6): 191:1–191:16. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/tog/YuZNDZ22,
author = {Zihan Yu and
Cheng Zhang and
Derek Nowrouzezahrai and
Zhao Dong and
Shuang Zhao},
title = {Efficient Differentiation of Pixel Reconstruction Filters for Path-Space
Differentiable Rendering},
journal = {{ACM} Trans. Graph.},
volume = {41},
number = {6},
pages = {191:1--191:16},
year = {2022},
url = {https://doi.org/10.1145/3550454.3555500},
doi = {10.1145/3550454.3555500},
timestamp = {Sun, 04 Aug 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/YuZNDZ22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Monte Carlo Method for Fluid Simulation.
Rioux-Lavoie, D.; Sugimoto, R.; Özdemir, T.; Shimada, N. H.; Batty, C.; Nowrouzezahrai, D.; and Hachisuka, T.
ACM Trans. Graph., 41(6): 240:1–240:16. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/tog/Rioux-LavoieSOS22,
author = {Damien Rioux{-}Lavoie and
Ryusuke Sugimoto and
T{\"{u}}may {\"{O}}zdemir and
Naoharu H. Shimada and
Christopher Batty and
Derek Nowrouzezahrai and
Toshiya Hachisuka},
title = {A Monte Carlo Method for Fluid Simulation},
journal = {{ACM} Trans. Graph.},
volume = {41},
number = {6},
pages = {240:1--240:16},
year = {2022},
url = {https://doi.org/10.1145/3550454.3555450},
doi = {10.1145/3550454.3555450},
timestamp = {Tue, 21 Mar 2023 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/Rioux-LavoieSOS22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
OSM: An Open Set Matting Framework with OOD Detection and Few-Shot Learning.
Zhou, Y.; Laradji, I. H.; Zhou, L.; and Nowrouzezahrai, D.
In 33rd British Machine Vision Conference 2022, BMVC 2022, London, UK, November 21-24, 2022, pages 92, 2022. BMVA Press
Paper
link
bibtex
@inproceedings{DBLP:conf/bmvc/ZhouLZN22,
author = {Yuhongze Zhou and
Issam Hadj Laradji and
Liguang Zhou and
Derek Nowrouzezahrai},
title = {{OSM:} An Open Set Matting Framework with {OOD} Detection and Few-Shot
Learning},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London,
UK, November 21-24, 2022},
pages = {92},
publisher = {{BMVA} Press},
year = {2022},
url = {https://bmvc2022.mpi-inf.mpg.de/92/},
timestamp = {Thu, 16 Feb 2023 16:15:04 +0100},
biburl = {https://dblp.org/rec/conf/bmvc/ZhouLZN22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Overcoming challenges in leveraging GANs for few-shot data augmentation.
Beckham, C.; Laradji, I. H.; Rodríguez, P.; Vázquez, D.; Nowrouzezahrai, D.; and Pal, C.
In Chandar, S.; Pascanu, R.; and Precup, D., editor(s), Conference on Lifelong Learning Agents, CoLLAs 2022, 22-24 August 2022, McGill University, Montréal, Québec, Canada, volume 199, of Proceedings of Machine Learning Research, pages 255–280, 2022. PMLR
Paper
link
bibtex
@inproceedings{DBLP:conf/collas/BeckhamLRVNP22,
author = {Christopher Beckham and
Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
David V{\'{a}}zquez and
Derek Nowrouzezahrai and
Christopher Pal},
editor = {Sarath Chandar and
Razvan Pascanu and
Doina Precup},
title = {Overcoming challenges in leveraging GANs for few-shot data augmentation},
booktitle = {Conference on Lifelong Learning Agents, CoLLAs 2022, 22-24 August
2022, McGill University, Montr{\'{e}}al, Qu{\'{e}}bec, Canada},
series = {Proceedings of Machine Learning Research},
volume = {199},
pages = {255--280},
publisher = {{PMLR}},
year = {2022},
url = {https://proceedings.mlr.press/v199/beckham22a.html},
timestamp = {Sat, 18 Feb 2023 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/collas/BeckhamLRVNP22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Kubric: A scalable dataset generator.
Greff, K.; Belletti, F.; Beyer, L.; Doersch, C.; Du, Y.; Duckworth, D.; Fleet, D. J.; Gnanapragasam, D.; Golemo, F.; Herrmann, C.; Kipf, T.; Kundu, A.; Lagun, D.; Laradji, I. H.; Liu, H. D.; Meyer, H.; Miao, Y.; Nowrouzezahrai, D.; Öztireli, A. C.; Pot, E.; Radwan, N.; Rebain, D.; Sabour, S.; Sajjadi, M. S. M.; Sela, M.; Sitzmann, V.; Stone, A.; Sun, D.; Vora, S.; Wang, Z.; Wu, T.; Yi, K. M.; Zhong, F.; and Tagliasacchi, A.
In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022, pages 3739–3751, 2022. IEEE
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/cvpr/GreffBBDDDFGGHK22,
author = {Klaus Greff and
Francois Belletti and
Lucas Beyer and
Carl Doersch and
Yilun Du and
Daniel Duckworth and
David J. Fleet and
Dan Gnanapragasam and
Florian Golemo and
Charles Herrmann and
Thomas Kipf and
Abhijit Kundu and
Dmitry Lagun and
Issam H. Laradji and
Hsueh{-}Ti Derek Liu and
Henning Meyer and
Yishu Miao and
Derek Nowrouzezahrai and
A. Cengiz {\"{O}}ztireli and
Etienne Pot and
Noha Radwan and
Daniel Rebain and
Sara Sabour and
Mehdi S. M. Sajjadi and
Matan Sela and
Vincent Sitzmann and
Austin Stone and
Deqing Sun and
Suhani Vora and
Ziyu Wang and
Tianhao Wu and
Kwang Moo Yi and
Fangcheng Zhong and
Andrea Tagliasacchi},
title = {Kubric: {A} scalable dataset generator},
booktitle = {{IEEE/CVF} Conference on Computer Vision and Pattern Recognition,
{CVPR} 2022, New Orleans, LA, USA, June 18-24, 2022},
pages = {3739--3751},
publisher = {{IEEE}},
year = {2022},
url = {https://doi.org/10.1109/CVPR52688.2022.00373},
doi = {10.1109/CVPR52688.2022.00373},
timestamp = {Mon, 18 Mar 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/cvpr/GreffBBDDDFGGHK22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learning to Guide and to be Guided in the Architect-Builder Problem.
Barde, P.; Karch, T.; Nowrouzezahrai, D.; Moulin-Frier, C.; Pal, C.; and Oudeyer, P.
In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022, 2022. OpenReview.net
Paper
link
bibtex
@inproceedings{DBLP:conf/iclr/BardeKNMPO22,
author = {Paul Barde and
Tristan Karch and
Derek Nowrouzezahrai and
Cl{\'{e}}ment Moulin{-}Frier and
Christopher Pal and
Pierre{-}Yves Oudeyer},
title = {Learning to Guide and to be Guided in the Architect-Builder Problem},
booktitle = {The Tenth International Conference on Learning Representations, {ICLR}
2022, Virtual Event, April 25-29, 2022},
publisher = {OpenReview.net},
year = {2022},
url = {https://openreview.net/forum?id=swiyAeGzFhQ},
timestamp = {Sat, 20 Aug 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/iclr/BardeKNMPO22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Attention-based Neural Cellular Automata.
Tesfaldet, M.; Nowrouzezahrai, D.; and Pal, C.
In Koyejo, S.; Mohamed, S.; Agarwal, A.; Belgrave, D.; Cho, K.; and Oh, A., editor(s), Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 - December 9, 2022, 2022.
Paper
link
bibtex
@inproceedings{DBLP:conf/nips/TesfaldetNP22,
author = {Mattie Tesfaldet and
Derek Nowrouzezahrai and
Chris Pal},
editor = {Sanmi Koyejo and
S. Mohamed and
A. Agarwal and
Danielle Belgrave and
K. Cho and
A. Oh},
title = {Attention-based Neural Cellular Automata},
booktitle = {Advances in Neural Information Processing Systems 35: Annual Conference
on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans,
LA, USA, November 28 - December 9, 2022},
year = {2022},
url = {http://papers.nips.cc/paper\_files/paper/2022/hash/361e5112d2eca09513bbd266e4b2d2be-Abstract-Conference.html},
timestamp = {Mon, 08 Jan 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/nips/TesfaldetNP22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Neural Shadow Mapping.
Datta, S.; Nowrouzezahrai, D.; Schied, C.; and Dong, Z.
In Nandigjav, M.; Mitra, N. J.; and Hertzmann, A., editor(s), SIGGRAPH '22: Special Interest Group on Computer Graphics and Interactive Techniques Conference, Vancouver, BC, Canada, August 7 - 11, 2022, pages 8:1–8:9, 2022. ACM
Paper
doi
link
bibtex
1 download
@inproceedings{DBLP:conf/siggraph/DattaNSD22,
author = {Sayantan Datta and
Derek Nowrouzezahrai and
Christoph Schied and
Zhao Dong},
editor = {Munkhtsetseg Nandigjav and
Niloy J. Mitra and
Aaron Hertzmann},
title = {Neural Shadow Mapping},
booktitle = {{SIGGRAPH} '22: Special Interest Group on Computer Graphics and Interactive
Techniques Conference, Vancouver, BC, Canada, August 7 - 11, 2022},
pages = {8:1--8:9},
publisher = {{ACM}},
year = {2022},
url = {https://doi.org/10.1145/3528233.3530700},
doi = {10.1145/3528233.3530700},
timestamp = {Mon, 05 Feb 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/siggraph/DattaNSD22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Compact Poisson Filters for Fast Fluid Simulation.
Rabbani, A. H.; Guertin, J.; Rioux-Lavoie, D.; Schoentgen, A.; Tong, K.; Sirois-Vigneux, A.; and Nowrouzezahrai, D.
In Nandigjav, M.; Mitra, N. J.; and Hertzmann, A., editor(s), SIGGRAPH '22: Special Interest Group on Computer Graphics and Interactive Techniques Conference, Vancouver, BC, Canada, August 7 - 11, 2022, pages 35:1–35:9, 2022. ACM
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/siggraph/RabbaniGRSTSN22,
author = {Amir Hossein Rabbani and
Jean{-}Philippe Guertin and
Damien Rioux{-}Lavoie and
Arnaud Schoentgen and
Kaitai Tong and
Alexandre Sirois{-}Vigneux and
Derek Nowrouzezahrai},
editor = {Munkhtsetseg Nandigjav and
Niloy J. Mitra and
Aaron Hertzmann},
title = {Compact Poisson Filters for Fast Fluid Simulation},
booktitle = {{SIGGRAPH} '22: Special Interest Group on Computer Graphics and Interactive
Techniques Conference, Vancouver, BC, Canada, August 7 - 11, 2022},
pages = {35:1--35:9},
publisher = {{ACM}},
year = {2022},
url = {https://doi.org/10.1145/3528233.3530737},
doi = {10.1145/3528233.3530737},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/siggraph/RabbaniGRSTSN22.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Kubric: A scalable dataset generator.
Greff, K.; Belletti, F.; Beyer, L.; Doersch, C.; Du, Y.; Duckworth, D.; Fleet, D. J.; Gnanapragasam, D.; Golemo, F.; Herrmann, C.; Kipf, T.; Kundu, A.; Lagun, D.; Laradji, I. H.; Liu, H. D.; Meyer, H.; Miao, Y.; Nowrouzezahrai, D.; Öztireli, C.; Pot, E.; Radwan, N.; Rebain, D.; Sabour, S.; Sajjadi, M. S. M.; Sela, M.; Sitzmann, V.; Stone, A.; Sun, D.; Vora, S.; Wang, Z.; Wu, T.; Yi, K. M.; Zhong, F.; and Tagliasacchi, A.
CoRR, abs/2203.03570. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2203-03570,
author = {Klaus Greff and
Francois Belletti and
Lucas Beyer and
Carl Doersch and
Yilun Du and
Daniel Duckworth and
David J. Fleet and
Dan Gnanapragasam and
Florian Golemo and
Charles Herrmann and
Thomas Kipf and
Abhijit Kundu and
Dmitry Lagun and
Issam H. Laradji and
Hsueh{-}Ti Derek Liu and
Henning Meyer and
Yishu Miao and
Derek Nowrouzezahrai and
Cengiz {\"{O}}ztireli and
Etienne Pot and
Noha Radwan and
Daniel Rebain and
Sara Sabour and
Mehdi S. M. Sajjadi and
Matan Sela and
Vincent Sitzmann and
Austin Stone and
Deqing Sun and
Suhani Vora and
Ziyu Wang and
Tianhao Wu and
Kwang Moo Yi and
Fangcheng Zhong and
Andrea Tagliasacchi},
title = {Kubric: {A} scalable dataset generator},
journal = {CoRR},
volume = {abs/2203.03570},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2203.03570},
doi = {10.48550/ARXIV.2203.03570},
eprinttype = {arXiv},
eprint = {2203.03570},
timestamp = {Mon, 18 Mar 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2203-03570.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Challenges in leveraging GANs for few-shot data augmentation.
Beckham, C.; Laradji, I. H.; Rodríguez, P.; Vázquez, D.; Nowrouzezahrai, D.; and Pal, C. J.
CoRR, abs/2203.16662. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2203-16662,
author = {Christopher Beckham and
Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
David V{\'{a}}zquez and
Derek Nowrouzezahrai and
Christopher J. Pal},
title = {Challenges in leveraging GANs for few-shot data augmentation},
journal = {CoRR},
volume = {abs/2203.16662},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2203.16662},
doi = {10.48550/ARXIV.2203.16662},
eprinttype = {arXiv},
eprint = {2203.16662},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2203-16662.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Latent Variable Models for Bayesian Causal Discovery.
Subramanian, J.; Annadani, Y.; Sheth, I.; Bauer, S.; Nowrouzezahrai, D.; and Kahou, S. E.
CoRR, abs/2207.05723. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2207-05723,
author = {Jithendaraa Subramanian and
Yashas Annadani and
Ivaxi Sheth and
Stefan Bauer and
Derek Nowrouzezahrai and
Samira Ebrahimi Kahou},
title = {Latent Variable Models for Bayesian Causal Discovery},
journal = {CoRR},
volume = {abs/2207.05723},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2207.05723},
doi = {10.48550/ARXIV.2207.05723},
eprinttype = {arXiv},
eprint = {2207.05723},
timestamp = {Thu, 14 Jul 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2207-05723.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Uncertainty-Driven Active Vision for Implicit Scene Reconstruction.
Smith, E. J.; Drozdzal, M.; Nowrouzezahrai, D.; Meger, D.; and Romero-Soriano, A.
CoRR, abs/2210.00978. 2022.
Paper
doi
link
bibtex
1 download
@article{DBLP:journals/corr/abs-2210-00978,
author = {Edward J. Smith and
Michal Drozdzal and
Derek Nowrouzezahrai and
David Meger and
Adriana Romero{-}Soriano},
title = {Uncertainty-Driven Active Vision for Implicit Scene Reconstruction},
journal = {CoRR},
volume = {abs/2210.00978},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2210.00978},
doi = {10.48550/ARXIV.2210.00978},
eprinttype = {arXiv},
eprint = {2210.00978},
timestamp = {Sun, 06 Oct 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2210-00978.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learning Latent Structural Causal Models.
Subramanian, J.; Annadani, Y.; Sheth, I.; Ke, N. R.; Deleu, T.; Bauer, S.; Nowrouzezahrai, D.; and Kahou, S. E.
CoRR, abs/2210.13583. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2210-13583,
author = {Jithendaraa Subramanian and
Yashas Annadani and
Ivaxi Sheth and
Nan Rosemary Ke and
Tristan Deleu and
Stefan Bauer and
Derek Nowrouzezahrai and
Samira Ebrahimi Kahou},
title = {Learning Latent Structural Causal Models},
journal = {CoRR},
volume = {abs/2210.13583},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2210.13583},
doi = {10.48550/ARXIV.2210.13583},
eprinttype = {arXiv},
eprint = {2210.13583},
timestamp = {Fri, 28 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2210-13583.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learning Neural Implicit Representations with Surface Signal Parameterizations.
Guan, Y.; Chubarau, A.; Rao, R.; and Nowrouzezahrai, D.
CoRR, abs/2211.00519. 2022.
Paper
doi
link
bibtex
2 downloads
@article{DBLP:journals/corr/abs-2211-00519,
author = {Yanran Guan and
Andrei Chubarau and
Ruby Rao and
Derek Nowrouzezahrai},
title = {Learning Neural Implicit Representations with Surface Signal Parameterizations},
journal = {CoRR},
volume = {abs/2211.00519},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2211.00519},
doi = {10.48550/ARXIV.2211.00519},
eprinttype = {arXiv},
eprint = {2211.00519},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2211-00519.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Attention-based Neural Cellular Automata.
Tesfaldet, M.; Nowrouzezahrai, D.; and Pal, C.
CoRR, abs/2211.01233. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2211-01233,
author = {Mattie Tesfaldet and
Derek Nowrouzezahrai and
Christopher Pal},
title = {Attention-based Neural Cellular Automata},
journal = {CoRR},
volume = {abs/2211.01233},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2211.01233},
doi = {10.48550/ARXIV.2211.01233},
eprinttype = {arXiv},
eprint = {2211.01233},
timestamp = {Fri, 04 Nov 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2211-01233.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Visual Question Answering From Another Perspective: CLEVR Mental Rotation Tests.
Beckham, C.; Weiss, M.; Golemo, F.; Honari, S.; Nowrouzezahrai, D.; and Pal, C.
CoRR, abs/2212.01639. 2022.
Paper
doi
link
bibtex
@article{DBLP:journals/corr/abs-2212-01639,
author = {Christopher Beckham and
Martin Weiss and
Florian Golemo and
Sina Honari and
Derek Nowrouzezahrai and
Christopher Pal},
title = {Visual Question Answering From Another Perspective: {CLEVR} Mental
Rotation Tests},
journal = {CoRR},
volume = {abs/2212.01639},
year = {2022},
url = {https://doi.org/10.48550/arXiv.2212.01639},
doi = {10.48550/ARXIV.2212.01639},
eprinttype = {arXiv},
eprint = {2212.01639},
timestamp = {Thu, 08 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2212-01639.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2021
(15)
|
Differentiable Compound Optics and Processing Pipeline Optimization for End-to-end Camera Design.
Tseng, E.; Mosleh, A.; Mannan, F.; Arnaud, K. S.; Sharma, A.; Peng, Y.; Braun, A.; Nowrouzezahrai, D.; Lalonde, J.; and Heide, F.
ACM Trans. Graph., 40(2): 18:1–18:19. 2021.
Paper
doi
link
bibtex
@article{DBLP:journals/tog/TsengMMASPBNLH21,
author = {Ethan Tseng and
Ali Mosleh and
Fahim Mannan and
Karl St. Arnaud and
Avinash Sharma and
Yifan Peng and
Alexander Braun and
Derek Nowrouzezahrai and
Jean{-}Fran{\c{c}}ois Lalonde and
Felix Heide},
title = {Differentiable Compound Optics and Processing Pipeline Optimization
for End-to-end Camera Design},
journal = {{ACM} Trans. Graph.},
volume = {40},
number = {2},
pages = {18:1--18:19},
year = {2021},
url = {https://doi.org/10.1145/3446791},
doi = {10.1145/3446791},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/TsengMMASPBNLH21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Neural Geometric Level of Detail: Real-Time Rendering With Implicit 3D Shapes.
Takikawa, T.; Litalien, J.; Yin, K.; Kreis, K.; Loop, C. T.; Nowrouzezahrai, D.; Jacobson, A.; McGuire, M.; and Fidler, S.
In IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021, pages 11358–11367, 2021. Computer Vision Foundation / IEEE
Paper
doi
link
bibtex
83 downloads
@inproceedings{DBLP:conf/cvpr/TakikawaLYKLNJM21,
author = {Towaki Takikawa and
Joey Litalien and
Kangxue Yin and
Karsten Kreis and
Charles T. Loop and
Derek Nowrouzezahrai and
Alec Jacobson and
Morgan McGuire and
Sanja Fidler},
title = {Neural Geometric Level of Detail: Real-Time Rendering With Implicit
3D Shapes},
booktitle = {{IEEE} Conference on Computer Vision and Pattern Recognition, {CVPR}
2021, virtual, June 19-25, 2021},
pages = {11358--11367},
publisher = {Computer Vision Foundation / {IEEE}},
year = {2021},
url = {https://openaccess.thecvf.com/content/CVPR2021/html/Takikawa\_Neural\_Geometric\_Level\_of\_Detail\_Real-Time\_Rendering\_With\_Implicit\_3D\_CVPR\_2021\_paper.html},
doi = {10.1109/CVPR46437.2021.01120},
timestamp = {Sun, 06 Oct 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/cvpr/TakikawaLYKLNJM21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
SSR: Semi-supervised Soft Rasterizer for single-view 2D to 3D Reconstruction.
Laradji, I. H.; Rodríguez, P.; Vázquez, D.; and Nowrouzezahrai, D.
In IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2021, Montreal, BC, Canada, October 11-17, 2021, pages 1427–1436, 2021. IEEE
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/iccvw/LaradjiRVN21,
author = {Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
David V{\'{a}}zquez and
Derek Nowrouzezahrai},
title = {{SSR:} Semi-supervised Soft Rasterizer for single-view 2D to 3D Reconstruction},
booktitle = {{IEEE/CVF} International Conference on Computer Vision Workshops,
{ICCVW} 2021, Montreal, BC, Canada, October 11-17, 2021},
pages = {1427--1436},
publisher = {{IEEE}},
year = {2021},
url = {https://doi.org/10.1109/ICCVW54120.2021.00164},
doi = {10.1109/ICCVW54120.2021.00164},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/iccvw/LaradjiRVN21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Regularized Inverse Reinforcement Learning.
Jeon, W.; Su, C.; Barde, P.; Doan, T.; Nowrouzezahrai, D.; and Pineau, J.
In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021, 2021. OpenReview.net
Paper
link
bibtex
@inproceedings{DBLP:conf/iclr/JeonSBDNP21,
author = {Wonseok Jeon and
Chen{-}Yang Su and
Paul Barde and
Thang Doan and
Derek Nowrouzezahrai and
Joelle Pineau},
title = {Regularized Inverse Reinforcement Learning},
booktitle = {9th International Conference on Learning Representations, {ICLR} 2021,
Virtual Event, Austria, May 3-7, 2021},
publisher = {OpenReview.net},
year = {2021},
url = {https://openreview.net/forum?id=HgLO8yalfwc},
timestamp = {Wed, 23 Jun 2021 17:36:39 +0200},
biburl = {https://dblp.org/rec/conf/iclr/JeonSBDNP21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
gradSim: Differentiable simulation for system identification and visuomotor control.
Murthy, J. K.; Macklin, M.; Golemo, F.; Voleti, V.; Petrini, L.; Weiss, M.; Considine, B.; Parent-Lévesque, J.; Xie, K.; Erleben, K.; Paull, L.; Shkurti, F.; Nowrouzezahrai, D.; and Fidler, S.
In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021, 2021. OpenReview.net
Paper
link
bibtex
@inproceedings{DBLP:conf/iclr/MurthyMGVPWCPXE21,
author = {J. Krishna Murthy and
Miles Macklin and
Florian Golemo and
Vikram Voleti and
Linda Petrini and
Martin Weiss and
Breandan Considine and
J{\'{e}}r{\^{o}}me Parent{-}L{\'{e}}vesque and
Kevin Xie and
Kenny Erleben and
Liam Paull and
Florian Shkurti and
Derek Nowrouzezahrai and
Sanja Fidler},
title = {gradSim: Differentiable simulation for system identification and visuomotor
control},
booktitle = {9th International Conference on Learning Representations, {ICLR} 2021,
Virtual Event, Austria, May 3-7, 2021},
publisher = {OpenReview.net},
year = {2021},
url = {https://openreview.net/forum?id=c\_E8kFWfhp0},
timestamp = {Tue, 06 Jul 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/iclr/MurthyMGVPWCPXE21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Dynamic Diffuse Global Illumination Resampling.
Majercik, Z.; Müller, T.; Keller, A.; Nowrouzezahrai, D.; and McGuire, M.
In SIGGRAPH 2021: Special Interest Group on Computer Graphics and Interactive Techniques Conference, Talks, Virtual Event, USA, August 9-13, 2021, pages 24:1–24:2, 2021. ACM
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/siggraph/Majercik00NM21,
author = {Zander Majercik and
Thomas M{\"{u}}ller and
Alexander Keller and
Derek Nowrouzezahrai and
Morgan McGuire},
title = {Dynamic Diffuse Global Illumination Resampling},
booktitle = {{SIGGRAPH} 2021: Special Interest Group on Computer Graphics and Interactive
Techniques Conference, Talks, Virtual Event, USA, August 9-13, 2021},
pages = {24:1--24:2},
publisher = {{ACM}},
year = {2021},
url = {https://doi.org/10.1145/3450623.3464635},
doi = {10.1145/3450623.3464635},
timestamp = {Tue, 10 Aug 2021 10:59:22 +0200},
biburl = {https://dblp.org/rec/conf/siggraph/Majercik00NM21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images.
Laradji, I. H.; Rodríguez, P.; Mañas, O.; Lensink, K.; Law, M.; Kurzman, L.; Parker, W.; Vázquez, D.; and Nowrouzezahrai, D.
In IEEE Winter Conference on Applications of Computer Vision, WACV 2021, Waikoloa, HI, USA, January 3-8, 2021, pages 2452–2461, 2021. IEEE
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/wacv/LaradjiRMLLKP0N21,
author = {Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
Oscar Ma{\~{n}}as and
Keegan Lensink and
Marco Law and
Lironne Kurzman and
William Parker and
David V{\'{a}}zquez and
Derek Nowrouzezahrai},
title = {A Weakly Supervised Consistency-based Learning Method for {COVID-19}
Segmentation in {CT} Images},
booktitle = {{IEEE} Winter Conference on Applications of Computer Vision, {WACV}
2021, Waikoloa, HI, USA, January 3-8, 2021},
pages = {2452--2461},
publisher = {{IEEE}},
year = {2021},
url = {https://doi.org/10.1109/WACV48630.2021.00250},
doi = {10.1109/WACV48630.2021.00250},
timestamp = {Wed, 07 Dec 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/wacv/LaradjiRMLLKP0N21.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes.
Takikawa, T.; Litalien, J.; Yin, K.; Kreis, K.; Loop, C. T.; Nowrouzezahrai, D.; Jacobson, A.; McGuire, M.; and Fidler, S.
CoRR, abs/2101.10994. 2021.
Paper
link
bibtex
83 downloads
@article{DBLP:journals/corr/abs-2101-10994,
author = {Towaki Takikawa and
Joey Litalien and
Kangxue Yin and
Karsten Kreis and
Charles T. Loop and
Derek Nowrouzezahrai and
Alec Jacobson and
Morgan McGuire and
Sanja Fidler},
title = {Neural Geometric Level of Detail: Real-time Rendering with Implicit
3D Shapes},
journal = {CoRR},
volume = {abs/2101.10994},
year = {2021},
url = {https://arxiv.org/abs/2101.10994},
eprinttype = {arXiv},
eprint = {2101.10994},
timestamp = {Thu, 19 May 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2101-10994.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Robust Motion In-betweening.
Harvey, F. G.; Yurick, M.; Nowrouzezahrai, D.; and Pal, C. J.
CoRR, abs/2102.04942. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2102-04942,
author = {F{\'{e}}lix G. Harvey and
Mike Yurick and
Derek Nowrouzezahrai and
Christopher J. Pal},
title = {Robust Motion In-betweening},
journal = {CoRR},
volume = {abs/2102.04942},
year = {2021},
url = {https://arxiv.org/abs/2102.04942},
eprinttype = {arXiv},
eprint = {2102.04942},
timestamp = {Thu, 18 Feb 2021 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2102-04942.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Countering Racial Bias in Computer Graphics Research.
Kim, T.; Rushmeier, H. E.; Dorsey, J.; Nowrouzezahrai, D.; Syed, R.; Jarosz, W.; and Darke, A. M.
CoRR, abs/2103.15163. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2103-15163,
author = {Theodore Kim and
Holly E. Rushmeier and
Julie Dorsey and
Derek Nowrouzezahrai and
Raqi Syed and
Wojciech Jarosz and
A. M. Darke},
title = {Countering Racial Bias in Computer Graphics Research},
journal = {CoRR},
volume = {abs/2103.15163},
year = {2021},
url = {https://arxiv.org/abs/2103.15163},
eprinttype = {arXiv},
eprint = {2103.15163},
timestamp = {Wed, 07 Apr 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2103-15163.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
gradSim: Differentiable simulation for system identification and visuomotor control.
Jatavallabhula, K. M.; Macklin, M.; Golemo, F.; Voleti, V.; Petrini, L.; Weiss, M.; Considine, B.; Parent-Lévesque, J.; Xie, K.; Erleben, K.; Paull, L.; Shkurti, F.; Nowrouzezahrai, D.; and Fidler, S.
CoRR, abs/2104.02646. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2104-02646,
author = {Krishna Murthy Jatavallabhula and
Miles Macklin and
Florian Golemo and
Vikram Voleti and
Linda Petrini and
Martin Weiss and
Breandan Considine and
J{\'{e}}r{\^{o}}me Parent{-}L{\'{e}}vesque and
Kevin Xie and
Kenny Erleben and
Liam Paull and
Florian Shkurti and
Derek Nowrouzezahrai and
Sanja Fidler},
title = {gradSim: Differentiable simulation for system identification and visuomotor
control},
journal = {CoRR},
volume = {abs/2104.02646},
year = {2021},
url = {https://arxiv.org/abs/2104.02646},
eprinttype = {arXiv},
eprint = {2104.02646},
timestamp = {Tue, 13 Apr 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-02646.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Dynamic Diffuse Global Illumination Resampling.
Majercik, Z.; Müller, T.; Keller, A.; Nowrouzezahrai, D.; and McGuire, M.
CoRR, abs/2108.05263. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2108-05263,
author = {Zander Majercik and
Thomas M{\"{u}}ller and
Alexander Keller and
Derek Nowrouzezahrai and
Morgan McGuire},
title = {Dynamic Diffuse Global Illumination Resampling},
journal = {CoRR},
volume = {abs/2108.05263},
year = {2021},
url = {https://arxiv.org/abs/2108.05263},
eprinttype = {arXiv},
eprint = {2108.05263},
timestamp = {Wed, 18 Aug 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2108-05263.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
SSR: Semi-supervised Soft Rasterizer for single-view 2D to 3D Reconstruction.
Laradji, I. H.; Rodríguez, P.; Vázquez, D.; and Nowrouzezahrai, D.
CoRR, abs/2108.09593. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2108-09593,
author = {Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
David V{\'{a}}zquez and
Derek Nowrouzezahrai},
title = {{SSR:} Semi-supervised Soft Rasterizer for single-view 2D to 3D Reconstruction},
journal = {CoRR},
volume = {abs/2108.09593},
year = {2021},
url = {https://arxiv.org/abs/2108.09593},
eprinttype = {arXiv},
eprint = {2108.09593},
timestamp = {Mon, 30 Aug 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2108-09593.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learning to Guide and to Be Guided in the Architect-Builder Problem.
Barde, P.; Karch, T.; Nowrouzezahrai, D.; Moulin-Frier, C.; Pal, C. J.; and Oudeyer, P.
CoRR, abs/2112.07342. 2021.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2112-07342,
author = {Paul Barde and
Tristan Karch and
Derek Nowrouzezahrai and
Cl{\'{e}}ment Moulin{-}Frier and
Christopher J. Pal and
Pierre{-}Yves Oudeyer},
title = {Learning to Guide and to Be Guided in the Architect-Builder Problem},
journal = {CoRR},
volume = {abs/2112.07342},
year = {2021},
url = {https://arxiv.org/abs/2112.07342},
eprinttype = {arXiv},
eprint = {2112.07342},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2112-07342.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2020
(20)
|
An Efficient Transport Estimator for Complex Layered Materials.
Gamboa, L. E.; Gruson, A.; and Nowrouzezahrai, D.
Comput. Graph. Forum, 39(2): 363–371. 2020.
Best Paper Honourable Mention Award!
Paper
doi
link
bibtex
4 downloads
@article{DBLP:journals/cgf/GamboaGN20,
author = {Luis E. Gamboa and
Adrien Gruson and
Derek Nowrouzezahrai},
title = {An Efficient Transport Estimator for Complex Layered Materials},
journal = {Comput. Graph. Forum},
volume = {39},
number = {2},
pages = {363--371},
year = {2020},
url = {https://doi.org/10.1111/cgf.13936},
doi = {10.1111/CGF.13936},
timestamp = {Wed, 07 Apr 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/GamboaGN20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Pix2Shape: Towards Unsupervised Learning of 3D Scenes from Images Using a View-Based Representation.
Rajeswar, S.; Mannan, F.; Golemo, F.; Parent-Lévesque, J.; Vázquez, D.; Nowrouzezahrai, D.; and Courville, A. C.
Int. J. Comput. Vis., 128(10): 2478–2493. 2020.
Paper
doi
link
bibtex
abstract We infer and generate three-dimensional (3D) scene information from a single input image and without supervision. This problem is under-explored, with most prior work relying on supervision from, e.g., 3D ground-truth, multiple images of a scene, image silhouettes or key-points. We propose Pix2Shape, an approach to solve this problem with four components: (i) an encoder that infers the latent 3D representation from an image, (ii) a decoder that generates an explicit 2.5D surfel-based reconstruction of a scene from the latent code (iii) a differentiable renderer that synthesizes a 2D image from the surfel representation, and (iv) a critic network trained to discriminate between images generated by the decoder-renderer and those from a training distribution. Pix2Shape can generate complex 3D scenes that scale with the view-dependent on-screen resolution, unlike representations that capture world-space resolution, i.e., voxels or meshes. We show that Pix2Shape learns a consistent scene representation in its encoded latent space and that the decoder can then be applied to this latent representation in order to synthesize the scene from a novel viewpoint. We evaluate Pix2Shape with experiments on the ShapeNet dataset as well as on a novel benchmark we developed, called 3D-IQTT, to evaluate models based on their ability to enable 3d spatial reasoning. Qualitative and quantitative evaluation demonstrate Pix2Shape's ability to solve scene reconstruction, generation, and understanding tasks.
@article{DBLP:journals/ijcv/RajeswarMGPVNC20,
author = {Sai Rajeswar and
Fahim Mannan and
Florian Golemo and
J{\'{e}}r{\^{o}}me Parent{-}L{\'{e}}vesque and
David V{\'{a}}zquez and
Derek Nowrouzezahrai and
Aaron C. Courville},
title = {Pix2Shape: Towards Unsupervised Learning of 3D Scenes from Images
Using a View-Based Representation},
journal = {Int. J. Comput. Vis.},
volume = {128},
number = {10},
pages = {2478--2493},
year = {2020},
url = {https://doi.org/10.1007/s11263-020-01322-1},
doi = {10.1007/S11263-020-01322-1},
timestamp = {Fri, 14 May 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/ijcv/RajeswarMGPVNC20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Delayed Rejection Metropolis Light Transport.
Rioux-Lavoie, D.; Litalien, J.; Gruson, A.; Hachisuka, T.; and Nowrouzezahrai, D.
ACM Trans. Graph., 39(3): 26:1–26:14. 2020.
Paper
doi
link
bibtex
3 downloads
@article{DBLP:journals/tog/Rioux-LavoieLGH20,
author = {Damien Rioux{-}Lavoie and
Joey Litalien and
Adrien Gruson and
Toshiya Hachisuka and
Derek Nowrouzezahrai},
title = {Delayed Rejection Metropolis Light Transport},
journal = {{ACM} Trans. Graph.},
volume = {39},
number = {3},
pages = {26:1--26:14},
year = {2020},
url = {https://doi.org/10.1145/3388538},
doi = {10.1145/3388538},
timestamp = {Sun, 06 Oct 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/Rioux-LavoieLGH20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Directional sources and listeners in interactive sound propagation using reciprocal wave field coding.
Chaitanya, C. R. A.; Raghuvanshi, N.; Godin, K. W.; Zhang, Z.; Nowrouzezahrai, D.; and Snyder, J. M.
ACM Trans. Graph., 39(4): 44. 2020.
Paper
doi
link
bibtex
@article{DBLP:journals/tog/ChaitanyaRGZNS20,
author = {Chakravarty R. Alla Chaitanya and
Nikunj Raghuvanshi and
Keith W. Godin and
Zechen Zhang and
Derek Nowrouzezahrai and
John M. Snyder},
title = {Directional sources and listeners in interactive sound propagation
using reciprocal wave field coding},
journal = {{ACM} Trans. Graph.},
volume = {39},
number = {4},
pages = {44},
year = {2020},
url = {https://doi.org/10.1145/3386569.3392459},
doi = {10.1145/3386569.3392459},
timestamp = {Sun, 25 Jul 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/ChaitanyaRGZNS20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Robust motion in-betweening.
Harvey, F. G.; Yurick, M.; Nowrouzezahrai, D.; and Pal, C. J.
ACM Trans. Graph., 39(4): 60. 2020.
Paper
doi
link
bibtex
2 downloads
@article{DBLP:journals/tog/HarveyYNP20,
author = {F{\'{e}}lix G. Harvey and
Mike Yurick and
Derek Nowrouzezahrai and
Christopher J. Pal},
title = {Robust motion in-betweening},
journal = {{ACM} Trans. Graph.},
volume = {39},
number = {4},
pages = {60},
year = {2020},
url = {https://doi.org/10.1145/3386569.3392480},
doi = {10.1145/3386569.3392480},
timestamp = {Tue, 09 Mar 2021 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/HarveyYNP20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
VoronoiNet : General Functional Approximators with Local Support.
Williams, F.; Parent-Lévesque, J.; Nowrouzezahrai, D.; Panozzo, D.; Yi, K. M.; and Tagliasacchi, A.
In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR Workshops 2020, Seattle, WA, USA, June 14-19, 2020, pages 1069–1073, 2020. Computer Vision Foundation / IEEE
Paper
doi
link
bibtex
@inproceedings{DBLP:conf/cvpr/WilliamsPNPYT20,
author = {Francis Williams and
J{\'{e}}r{\^{o}}me Parent{-}L{\'{e}}vesque and
Derek Nowrouzezahrai and
Daniele Panozzo and
Kwang Moo Yi and
Andrea Tagliasacchi},
title = {VoronoiNet : General Functional Approximators with Local Support},
booktitle = {2020 {IEEE/CVF} Conference on Computer Vision and Pattern Recognition,
{CVPR} Workshops 2020, Seattle, WA, USA, June 14-19, 2020},
pages = {1069--1073},
publisher = {Computer Vision Foundation / {IEEE}},
year = {2020},
url = {https://openaccess.thecvf.com/content\_CVPRW\_2020/html/w17/Williams\_VoronoiNet\_General\_Functional\_Approximators\_With\_Local\_Support\_CVPRW\_2020\_paper.html},
doi = {10.1109/CVPRW50498.2020.00140},
timestamp = {Tue, 31 Aug 2021 14:00:09 +0200},
biburl = {https://dblp.org/rec/conf/cvpr/WilliamsPNPYT20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Using Speech Synthesis to Train End-To-End Spoken Language Understanding Models.
Lugosch, L.; Meyer, B. H.; Nowrouzezahrai, D.; and Ravanelli, M.
In 2020 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2020, Barcelona, Spain, May 4-8, 2020, pages 8499–8503, 2020. IEEE
Paper
doi
link
bibtex
1 download
@inproceedings{DBLP:conf/icassp/LugoschMNR20,
author = {Loren Lugosch and
Brett H. Meyer and
Derek Nowrouzezahrai and
Mirco Ravanelli},
title = {Using Speech Synthesis to Train End-To-End Spoken Language Understanding
Models},
booktitle = {2020 {IEEE} International Conference on Acoustics, Speech and Signal
Processing, {ICASSP} 2020, Barcelona, Spain, May 4-8, 2020},
pages = {8499--8503},
publisher = {{IEEE}},
year = {2020},
url = {https://doi.org/10.1109/ICASSP40776.2020.9053063},
doi = {10.1109/ICASSP40776.2020.9053063},
timestamp = {Thu, 23 Jul 2020 16:19:28 +0200},
biburl = {https://dblp.org/rec/conf/icassp/LugoschMNR20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization.
Barde, P.; Roy, J.; Jeon, W.; Pineau, J.; Pal, C.; and Nowrouzezahrai, D.
In Larochelle, H.; Ranzato, M.; Hadsell, R.; Balcan, M.; and Lin, H., editor(s), Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020.
Paper
link
bibtex
@inproceedings{DBLP:conf/nips/BardeRJPPN20,
author = {Paul Barde and
Julien Roy and
Wonseok Jeon and
Joelle Pineau and
Chris Pal and
Derek Nowrouzezahrai},
editor = {Hugo Larochelle and
Marc'Aurelio Ranzato and
Raia Hadsell and
Maria{-}Florina Balcan and
Hsuan{-}Tien Lin},
title = {Adversarial Soft Advantage Fitting: Imitation Learning without Policy
Optimization},
booktitle = {Advances in Neural Information Processing Systems 33: Annual Conference
on Neural Information Processing Systems 2020, NeurIPS 2020, December
6-12, 2020, virtual},
year = {2020},
url = {https://proceedings.neurips.cc/paper/2020/hash/9161ab7a1b61012c4c303f10b4c16b2c-Abstract.html},
timestamp = {Tue, 19 Jan 2021 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/nips/BardeRJPPN20.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Scalable Multi-Agent Inverse Reinforcement Learning via Actor-Attention-Critic.
Jeon, W.; Barde, P.; Nowrouzezahrai, D.; and Pineau, J.
CoRR, abs/2002.10525. 2020.
Published in the AAAI Workshop on Reinforcement Learning in Games (AAAI RLG)
Paper (PDF)
Paper
link
bibtex
abstract Multi-agent adversarial inverse reinforcement learning (MA-AIRL) is a recent approach that applies single-agent AIRL to multi-agent problems where we seek to recover both policies for our agents and reward functions that promote expert-like behavior. While MA-AIRL has promising results on cooperative and competitive tasks, it is sample-inefficient and has only been validated empirically for small numbers of agents – its ability to scale to many agents remains an open question. We propose a multi-agent inverse RL algorithm that is more sample-efficient and scalable than previous works. Specifically, we employ multi-agent actor-attention-critic (MAAC) – an off-policy multi-agent RL (MARL) method – for the RL inner loop of the inverse RL procedure. In doing so, we are able to increase sample efficiency compared to state-of-the-art baselines, across both small- and large-scale tasks. Moreover, the RL agents trained on the rewards recovered by our method better match the experts than those trained on the rewards derived from the baselines. Finally, our method requires far fewer agent-environment interactions, particularly as the number of agents increases.
@article{DBLP:journals/corr/abs-2002-10525,
author = {Wonseok Jeon and
Paul Barde and
Derek Nowrouzezahrai and
Joelle Pineau},
title = {Scalable Multi-Agent Inverse Reinforcement Learning via Actor-Attention-Critic},
journal = {CoRR},
volume = {abs/2002.10525},
year = {2020},
url = {https://arxiv.org/abs/2002.10525},
eprinttype = {arXiv},
eprint = {2002.10525},
timestamp = {Tue, 03 Mar 2020 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2002-10525.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Pix2Shape - Towards Unsupervised Learning of 3D Scenes from Images using a View-based Representation.
Rajeswar, S.; Mannan, F.; Golemo, F.; Parent-Lévesque, J.; Vázquez, D.; Nowrouzezahrai, D.; and Courville, A. C.
CoRR, abs/2003.14166. 2020.
Paper
link
bibtex
abstract We infer and generate three-dimensional (3D) scene information from a single input image and without supervision. This problem is under-explored, with most prior work relying on supervision from, e.g., 3D ground-truth, multiple images of a scene, image silhouettes or key-points. We propose Pix2Shape, an approach to solve this problem with four components: (i) an encoder that infers the latent 3D representation from an image, (ii) a decoder that generates an explicit 2.5D surfel-based reconstruction of a scene from the latent code (iii) a differentiable renderer that synthesizes a 2D image from the surfel representation, and (iv) a critic network trained to discriminate between images generated by the decoder-renderer and those from a training distribution. Pix2Shape can generate complex 3D scenes that scale with the view-dependent on-screen resolution, unlike representations that capture world-space resolution, i.e., voxels or meshes. We show that Pix2Shape learns a consistent scene representation in its encoded latent space and that the decoder can then be applied to this latent representation in order to synthesize the scene from a novel viewpoint. We evaluate Pix2Shape with experiments on the ShapeNet dataset as well as on a novel benchmark we developed, called 3D-IQTT, to evaluate models based on their ability to enable 3d spatial reasoning. Qualitative and quantitative evaluation demonstrate Pix2Shape's ability to solve scene reconstruction, generation, and understanding tasks.
@article{DBLP:journals/corr/abs-2003-14166,
author = {Sai Rajeswar and
Fahim Mannan and
Florian Golemo and
J{\'{e}}r{\^{o}}me Parent{-}L{\'{e}}vesque and
David V{\'{a}}zquez and
Derek Nowrouzezahrai and
Aaron C. Courville},
title = {Pix2Shape - Towards Unsupervised Learning of 3D Scenes from Images
using a View-based Representation},
journal = {CoRR},
volume = {abs/2003.14166},
year = {2020},
url = {https://arxiv.org/abs/2003.14166},
eprinttype = {arXiv},
eprint = {2003.14166},
timestamp = {Fri, 14 May 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2003-14166.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Surprisal-Triggered Conditional Computation with Neural Networks.
Lugosch, L.; Nowrouzezahrai, D.; and Meyer, B. H.
CoRR, abs/2006.01659. 2020.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2006-01659,
author = {Loren Lugosch and
Derek Nowrouzezahrai and
Brett H. Meyer},
title = {Surprisal-Triggered Conditional Computation with Neural Networks},
journal = {CoRR},
volume = {abs/2006.01659},
year = {2020},
url = {https://arxiv.org/abs/2006.01659},
eprinttype = {arXiv},
eprint = {2006.01659},
timestamp = {Mon, 08 Jun 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2006-01659.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization.
Barde, P.; Roy, J.; Jeon, W.; Pineau, J.; Pal, C. J.; and Nowrouzezahrai, D.
CoRR, abs/2006.13258. 2020.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2006-13258,
author = {Paul Barde and
Julien Roy and
Wonseok Jeon and
Joelle Pineau and
Christopher J. Pal and
Derek Nowrouzezahrai},
title = {Adversarial Soft Advantage Fitting: Imitation Learning without Policy
Optimization},
journal = {CoRR},
volume = {abs/2006.13258},
year = {2020},
url = {https://arxiv.org/abs/2006.13258},
eprinttype = {arXiv},
eprint = {2006.13258},
timestamp = {Wed, 01 Jul 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2006-13258.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images.
Laradji, I. H.; Rodríguez, P.; Mañas, O.; Lensink, K.; Law, M.; Kurzman, L.; Parker, W.; Vázquez, D.; and Nowrouzezahrai, D.
CoRR, abs/2007.02180. 2020.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2007-02180,
author = {Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
Oscar Ma{\~{n}}as and
Keegan Lensink and
Marco Law and
Lironne Kurzman and
William Parker and
David V{\'{a}}zquez and
Derek Nowrouzezahrai},
title = {A Weakly Supervised Consistency-based Learning Method for {COVID-19}
Segmentation in {CT} Images},
journal = {CoRR},
volume = {abs/2007.02180},
year = {2020},
url = {https://arxiv.org/abs/2007.02180},
eprinttype = {arXiv},
eprint = {2007.02180},
timestamp = {Fri, 14 May 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2007-02180.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Weakly Supervised Region-Based Active Learning Method for COVID-19 Segmentation in CT Images.
Laradji, I. H.; Rodríguez, P.; Branchaud-Charron, F.; Lensink, K.; Atighehchian, P.; Parker, W.; Vázquez, D.; and Nowrouzezahrai, D.
CoRR, abs/2007.07012. 2020.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2007-07012,
author = {Issam H. Laradji and
Pau Rodr{\'{\i}}guez and
Frederic Branchaud{-}Charron and
Keegan Lensink and
Parmida Atighehchian and
William Parker and
David V{\'{a}}zquez and
Derek Nowrouzezahrai},
title = {A Weakly Supervised Region-Based Active Learning Method for {COVID-19}
Segmentation in {CT} Images},
journal = {CoRR},
volume = {abs/2007.07012},
year = {2020},
url = {https://arxiv.org/abs/2007.07012},
eprinttype = {arXiv},
eprint = {2007.07012},
timestamp = {Fri, 14 May 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2007-07012.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Overfit Neural Networks as a Compact Shape Representation.
Davies, T.; Nowrouzezahrai, D.; and Jacobson, A.
CoRR, abs/2009.09808. 2020.
Paper
link
bibtex
1 download
@article{DBLP:journals/corr/abs-2009-09808,
author = {Thomas Davies and
Derek Nowrouzezahrai and
Alec Jacobson},
title = {Overfit Neural Networks as a Compact Shape Representation},
journal = {CoRR},
volume = {abs/2009.09808},
year = {2020},
url = {https://arxiv.org/abs/2009.09808},
eprinttype = {arXiv},
eprint = {2009.09808},
timestamp = {Wed, 23 Sep 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2009-09808.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Regularized Inverse Reinforcement Learning.
Jeon, W.; Su, C.; Barde, P.; Doan, T.; Nowrouzezahrai, D.; and Pineau, J.
CoRR, abs/2010.03691. 2020.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-2010-03691,
author = {Wonseok Jeon and
Chen{-}Yang Su and
Paul Barde and
Thang Doan and
Derek Nowrouzezahrai and
Joelle Pineau},
title = {Regularized Inverse Reinforcement Learning},
journal = {CoRR},
volume = {abs/2010.03691},
year = {2020},
url = {https://arxiv.org/abs/2010.03691},
eprinttype = {arXiv},
eprint = {2010.03691},
timestamp = {Tue, 13 Oct 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2010-03691.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Affinity LCFCN: Learning to Segment Fish with Weak Supervision.
Laradji, I. H.; Saleh, A.; Rodríguez, P.; Nowrouzezahrai, D.; Azghadi, M. R.; and Vázquez, D.
CoRR, abs/2011.03149. 2020.
Paper
link
bibtex
1 download
@article{DBLP:journals/corr/abs-2011-03149,
author = {Issam H. Laradji and
Alzayat Saleh and
Pau Rodr{\'{\i}}guez and
Derek Nowrouzezahrai and
Mostafa Rahimi Azghadi and
David V{\'{a}}zquez},
title = {Affinity {LCFCN:} Learning to Segment Fish with Weak Supervision},
journal = {CoRR},
volume = {abs/2011.03149},
year = {2020},
url = {https://arxiv.org/abs/2011.03149},
eprinttype = {arXiv},
eprint = {2011.03149},
timestamp = {Thu, 14 Oct 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2011-03149.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2019
(17)
|
A Survey on Gradient-Domain Rendering.
Hua, B.; Gruson, A.; Petitjean, V.; Zwicker, M.; Nowrouzezahrai, D.; Eisemann, E.; and Hachisuka, T.
Comput. Graph. Forum, 38(2): 455–472. 2019.
Presented at Eurographics
Paper (PDF)
Paper
doi
link
bibtex
abstract Monte Carlo methods for physically-based light transport simulation are broadly adopted in the feature film production, animation and visual effects industries. These methods, however, often result in noisy images and have slow convergence. As such, improving the convergence of Monte Carlo rendering remains an important open problem. Gradient-domain light transport is a recent family of techniques that can accelerate Monte Carlo rendering by up to an order of magnitude, leveraging a gradient-based estimation and a reformulation of the rendering problem as an image reconstruction. This state of the art report comprehensively frames the fundamentals of gradient-domain rendering, as well as the pragmatic details behind practical gradient-domain uni- and bidirectional path tracing and photon density estimation algorithms. Moreover, we discuss the various image reconstruction schemes that are crucial to accurate and stable gradient-domain rendering. Finally, we benchmark various gradient-domain techniques against the state-of-the-art in denoising methods before discussing open problems.
@article{DBLP:journals/cgf/HuaGPZNEH19,
author = {Binh{-}Son Hua and
Adrien Gruson and
Victor Petitjean and
Matthias Zwicker and
Derek Nowrouzezahrai and
Elmar Eisemann and
Toshiya Hachisuka},
title = {A Survey on Gradient-Domain Rendering},
journal = {Comput. Graph. Forum},
volume = {38},
number = {2},
pages = {455--472},
year = {2019},
url = {https://doi.org/10.1111/cgf.13652},
doi = {10.1111/CGF.13652},
timestamp = {Thu, 04 Jul 2019 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/HuaGPZNEH19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Volume Path Guiding Based on Zero-Variance Random Walk Theory.
Herholz, S.; Zhao, Y.; Elek, O.; Nowrouzezahrai, D.; Lensch, H. P. A.; and Krivánek, J.
ACM Trans. Graph., 38(3): 25:1–25:19. 2019.
Presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
abstract The efficiency of Monte Carlo methods, commonly used to render participating media, is directly linked to the manner in which random sampling decisions are made during path construction. Notably, path construction is influenced by scattering direction and distance sampling, Russian roulette, and splitting strategies. We present a consistent suite of volumetric path construction techniques where all these sampling decisions are guided by a cached estimate of the adjoint transport solution. The proposed strategy is based on the theory of zero-variance path sampling schemes, accounting for the spatial and directional variation in volumetric transport. Our key technical contribution, enabling the use of this approach in the context of volume light transport, is a novel guiding strategy for sampling the particle collision distance proportionally to the product of transmittance and the adjoint transport solution (e.g., in-scattered radiance). Furthermore, scattering directions are likewise sampled according to the product of the phase function and the incident radiance estimate. Combined with guided Russian roulette and splitting strategies tailored to volumes, we demonstrate about an order-of-magnitude error reduction compared to standard unidirectional methods. Consequently, our approach can render scenes otherwise intractable for such methods, while still retaining their simplicity (compared to, e.g., bidirectional methods).
@article{DBLP:journals/tog/HerholzZENLK19,
author = {Sebastian Herholz and
Yangyang Zhao and
Oskar Elek and
Derek Nowrouzezahrai and
Hendrik P. A. Lensch and
Jaroslav Kriv{\'{a}}nek},
title = {Volume Path Guiding Based on Zero-Variance Random Walk Theory},
journal = {{ACM} Trans. Graph.},
volume = {38},
number = {3},
pages = {25:1--25:19},
year = {2019},
url = {https://doi.org/10.1145/3230635},
doi = {10.1145/3230635},
timestamp = {Mon, 05 Feb 2024 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/HerholzZENLK19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Hyperparameter optimization in black-box image processing using differentiable proxies.
Tseng, E.; Yu, F.; Yang, Y.; Mannan, F.; Arnaud, K. S.; Nowrouzezahrai, D.; Lalonde, J.; and Heide, F.
ACM Trans. Graph., 38(4): 27:1–27:14. 2019.
Paper (PDF) Supplement (PDF)
Paper
doi
link
bibtex
abstract Nearly every commodity imaging system we directly interact with, or indirectly rely on, leverages power efficient, application-adjustable black-box hardware image signal processing (ISPs) units, running either in dedicated hardware blocks, or as proprietary software modules on programmable hardware. The configuration parameters of these black-box ISPs often have complex interactions with the output image, and must be adjusted prior to deployment according to application-specific quality and performance metrics. Today, this search is commonly performed manually by "golden eye" experts or algorithm developers leveraging domain expertise. We present a fully automatic system to optimize the parameters of black-box hardware and software image processing pipelines according to any arbitrary (i.e., application-specific) metric. We leverage a differentiable mapping between the configuration space and evaluation metrics, parameterized by a convolutional neural network that we train in an end-to-end fashion with imaging hardware in-the-loop. Unlike prior art, our differentiable proxies allow for high-dimension parameter search with stochastic first-order optimizers, without explicitly modeling any lower-level image processing transformations. As such, we can efficiently optimize black-box image processing pipelines for a variety of imaging applications, reducing application-specific configuration times from months to hours. Our optimization method is fully automatic, even with black-box hardware in the loop. We validate our method on experimental data for real-time display applications, object detection, and extreme low-light imaging. The proposed approach outperforms manual search qualitatively and quantitatively for all domain-specific applications tested. When applied to traditional denoisers, we demonstrate that — just by changing hyperparameters — traditional algorithms can outperform recent deep learning methods by a substantial margin on recent benchmarks.
@article{DBLP:journals/tog/TsengYYMANLH19,
author = {Ethan Tseng and
Felix Yu and
Yuting Yang and
Fahim Mannan and
Karl St. Arnaud and
Derek Nowrouzezahrai and
Jean{-}Fran{\c{c}}ois Lalonde and
Felix Heide},
title = {Hyperparameter optimization in black-box image processing using differentiable
proxies},
journal = {{ACM} Trans. Graph.},
volume = {38},
number = {4},
pages = {27:1--27:14},
year = {2019},
url = {https://doi.org/10.1145/3306346.3322996},
doi = {10.1145/3306346.3322996},
timestamp = {Thu, 14 Oct 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/TsengYYMANLH19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Adaptive Sampling for Sound Propagation.
Chaitanya, C. R. A.; Snyder, J. M.; Godin, K. W.; Nowrouzezahrai, D.; and Raghuvanshi, N.
IEEE Trans. Vis. Comput. Graph., 25(5): 1846–1854. 2019.
Presented at the IEEE Virtual Reality Conference
Paper (PDF)
Paper
doi
link
bibtex
abstract Precomputed sound propagation samples acoustics at discrete scene probe positions to support dynamic listener locations. An offline 3D numerical simulation is performed at each probe and the resulting field is encoded for runtime rendering with dynamic sources. Prior work place probes on a uniform grid, requiring high density to resolve narrow spaces. Our adaptive sampling approach varies probe density based on a novel "local diameter" measure of the space surrounding a given point, evaluated by stochastically tracing paths in the scene. We apply this measure to layout probes so as to smoothly adapt resolution and eliminate undersampling in corners, narrow corridors and stairways, while coarsening appropriately in more open areas. Coupled with a new runtime interpolator based on radial weights over geodesic paths, we achieve smooth acoustic effects that respect scene boundaries as both the source or listener move, unlike existing visibility-based solutions. We consistently demonstrate quality improvement over prior work at fixed cost.
@article{DBLP:journals/tvcg/ChaitanyaSGNR19,
author = {Chakravarty R. Alla Chaitanya and
John M. Snyder and
Keith W. Godin and
Derek Nowrouzezahrai and
Nikunj Raghuvanshi},
title = {Adaptive Sampling for Sound Propagation},
journal = {{IEEE} Trans. Vis. Comput. Graph.},
volume = {25},
number = {5},
pages = {1846--1854},
year = {2019},
url = {https://doi.org/10.1109/TVCG.2019.2898765},
doi = {10.1109/TVCG.2019.2898765},
timestamp = {Mon, 26 Oct 2020 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tvcg/ChaitanyaSGNR19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Navigation Agents for the Visually Impaired: A Sidewalk Simulator and Experiments.
Weiss, M.; Chamorro, S.; Girgis, R.; Luck, M.; Kahou, S. E.; Cohen, J. P.; Nowrouzezahrai, D.; Precup, D.; Golemo, F.; and Pal, C.
In Kaelbling, L. P.; Kragic, D.; and Sugiura, K., editor(s), 3rd Annual Conference on Robot Learning, CoRL 2019, Osaka, Japan, October 30 - November 1, 2019, Proceedings, volume 100, of Proceedings of Machine Learning Research, pages 1314–1327, 2019. PMLR
Paper
link
bibtex
abstract Millions of blind and visually-impaired (BVI) people navigate urban environments every day, using smartphones for high-level path-planning and white canes or guide dogs for local information. However, many BVI people still struggle to travel to new places. In our endeavor to create a navigation assistant for the BVI, we found that existing Reinforcement Learning (RL) environments were unsuitable for the task. This work introduces SEVN, a sidewalk simulation environment and a neural network-based approach to creating a navigation agent. SEVN contains panoramic images with labels for house numbers, doors, and street name signs, and formulations for several navigation tasks. We study the performance of an RL algorithm (PPO) in this setting. Our policy model fuses multi-modal observations in the form of variable resolution images, visible text, and simulated GPS data to navigate to a goal door. We hope that this dataset, simulator, and experimental results will provide a foundation for further research into the creation of agents that can assist members of the BVI community with outdoor navigation.
@inproceedings{DBLP:conf/corl/WeissCGLKCNPGP19,
author = {Martin Weiss and
Simon Chamorro and
Roger Girgis and
Margaux Luck and
Samira Ebrahimi Kahou and
Joseph Paul Cohen and
Derek Nowrouzezahrai and
Doina Precup and
Florian Golemo and
Chris Pal},
editor = {Leslie Pack Kaelbling and
Danica Kragic and
Komei Sugiura},
title = {Navigation Agents for the Visually Impaired: {A} Sidewalk Simulator
and Experiments},
booktitle = {3rd Annual Conference on Robot Learning, CoRL 2019, Osaka, Japan,
October 30 - November 1, 2019, Proceedings},
series = {Proceedings of Machine Learning Research},
volume = {100},
pages = {1314--1327},
publisher = {{PMLR}},
year = {2019},
url = {http://proceedings.mlr.press/v100/weiss20a.html},
timestamp = {Mon, 25 May 2020 12:12:52 +0200},
biburl = {https://dblp.org/rec/conf/corl/WeissCGLKCNPGP19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Frequency Analysis and Dual Hierarchy for Efficient Rendering of Subsurface Scattering.
Milaenen, D.; Belcour, L.; Guertin, J.; Hachisuka, T.; and Nowrouzezahrai, D.
In Tagliasacchi, A.; and Teather, R. J., editor(s), Proceedings of the 45th Graphics Interface Conference 2019, Kingston, Ontario, Canada, May 28-31, 2019, pages 3:1–3:7, 2019. Canadian Human-Computer Communications Society / ACM
Best Paper Award Winner!
Paper (PDF)
Paper
link
bibtex
abstract BSSRDFs are commonly used to model subsurface light transport in highly scattering media such as skin and marble. Rendering with BSSRDFs requires an additional spatial integration, which can be significantly more expensive than surface-only rendering with BRDFs. We introduce a novel hierarchical rendering method that can mitigate this additional spatial integration cost. Our method has two key components: a novel frequency analysis of subsurface light transport, and a dual hierarchy over shading and illumination samples. Our frequency analysis predicts the spatial and angular variation of outgoing radiance due to a BSSRDF. We use this analysis to drive adaptive spatial BSSRDF integration with sparse image and illumination samples. We propose the use of a dual-tree structure that allows us to simultaneously traverse a tree of shade points (i.e., pixels) and a tree of object-space illumination samples. Our dual- tree approach generalizes existing single-tree accelerations. Both our frequency analysis and the dual-tree structure are compatible with most existing BSSRDF models, and we show that our method improves rendering times compared to the state of the art method of Jensen and Buhler.
@inproceedings{DBLP:conf/graphicsinterface/MilaenenBGHN19,
author = {David Milaenen and
Laurent Belcour and
Jean{-}Philippe Guertin and
Toshiya Hachisuka and
Derek Nowrouzezahrai},
editor = {Andrea Tagliasacchi and
Robert J. Teather},
title = {A Frequency Analysis and Dual Hierarchy for Efficient Rendering of
Subsurface Scattering},
booktitle = {Proceedings of the 45th Graphics Interface Conference 2019, Kingston,
Ontario, Canada, May 28-31, 2019},
pages = {3:1--3:7},
publisher = {Canadian Human-Computer Communications Society / {ACM}},
year = {2019},
url = {https://dl.acm.org/citation.cfm?id=3371604},
timestamp = {Fri, 08 Nov 2019 13:41:09 +0100},
biburl = {https://dblp.org/rec/conf/graphicsinterface/MilaenenBGHN19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
View-dependent Radiance Caching.
Zhao, Y.; Belcour, L.; and Nowrouzezahrai, D.
In Tagliasacchi, A.; and Teather, R. J., editor(s), Proceedings of the 45th Graphics Interface Conference 2019, Kingston, Ontario, Canada, May 28-31, 2019, pages 22:1–22:9, 2019. Canadian Human-Computer Communications Society / ACM
Paper (PDF)
Paper
link
bibtex
abstract Radiance caching is used to accelerate global illumination computations, exploiting the spatial coherence of indirect illumination on surfaces. We propose a new radiance caching approach capable of more correctly reconstructing inter-reflections between glossy surfaces, all while improving performance compared to previous approaches. Contrary to previous works, our view-dependent radiance caching scheme does not heavily rely fundamentally on basis-space representations such as spherical harmonics, and can directly treat outgoing radiance at surfaces instead of incoming radiance distributions. We introduce a new view-dependent record placement strategy and adapt recent Hessian-based error metrics to our view-dependent records. To do so, we derive and compute more accurate derivatives of radiance at surfaces in the scene.
@inproceedings{DBLP:conf/graphicsinterface/ZhaoBN19,
author = {Yangyang Zhao and
Laurent Belcour and
Derek Nowrouzezahrai},
editor = {Andrea Tagliasacchi and
Robert J. Teather},
title = {View-dependent Radiance Caching},
booktitle = {Proceedings of the 45th Graphics Interface Conference 2019, Kingston,
Ontario, Canada, May 28-31, 2019},
pages = {22:1--22:9},
publisher = {Canadian Human-Computer Communications Society / {ACM}},
year = {2019},
url = {https://dl.acm.org/citation.cfm?id=3371630},
timestamp = {Fri, 08 Nov 2019 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/graphicsinterface/ZhaoBN19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Beyond Pixel Norm-Balls: Parametric Adversaries using an Analytically Differentiable Renderer.
Liu, H. D.; Tao, M.; Li, C.; Nowrouzezahrai, D.; and Jacobson, A.
In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019, 2019. OpenReview.net
Paper (PDF)
Paper
link
bibtex
abstract Many machine learning image classifiers are vulnerable to adversarial attacks, inputs with perturbations designed to intentionally trigger misclassification. Current adversarial methods directly alter pixel colors and evaluate against pixel norm-balls: pixel perturbations smaller than a specified magnitude, according to a measurement norm. This evaluation, however, has limited practical utility since perturbations in the pixel space do not correspond to underlying real-world phenomena of image formation that lead to them and has no security motivation attached. Pixels in natural images are measurements of light that has interacted with the geometry of a physical scene. As such, we propose a novel evaluation measure, parametric norm-balls, by directly perturbing physical parameters that underly image formation. One enabling contribution we present is a physically-based differentiable renderer that allows us to propagate pixel gradients to the parametric space of lighting and geometry. Our approach enables physically-based adversarial attacks, and our differentiable renderer leverages models from the interactive rendering literature to balance the performance and accuracy trade-offs necessary for a memory-efficient and scalable adversarial data augmentation workflow.
@inproceedings{DBLP:conf/iclr/LiuTLNJ19,
author = {Hsueh{-}Ti Derek Liu and
Michael Tao and
Chun{-}Liang Li and
Derek Nowrouzezahrai and
Alec Jacobson},
title = {Beyond Pixel Norm-Balls: Parametric Adversaries using an Analytically
Differentiable Renderer},
booktitle = {7th International Conference on Learning Representations, {ICLR} 2019,
New Orleans, LA, USA, May 6-9, 2019},
publisher = {OpenReview.net},
year = {2019},
url = {https://openreview.net/forum?id=SJl2niR9KQ},
timestamp = {Mon, 24 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/iclr/LiuTLNJ19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Subspace neural physics: fast data-driven interactive simulation.
Holden, D.; Duong, B. C.; Datta, S.; and Nowrouzezahrai, D.
In Lee, S.; Schroeder, C. A.; Spencer, S. N.; Batty, C.; and Huang, J., editor(s), Proceedings of the 18th annual ACM SIGGRAPH/Eurographics Symposium on Computer Animation, SCA 2019, Los Angeles, CA, USA, July 26-28, 2019, pages 6:1–6:12, 2019. ACM
Paper (PDF)
Paper
doi
link
bibtex
3 downloads
abstract Data-driven methods for physical simulation are an attractive option for interactive applications due to their ability to trade precomputation and memory footprint in exchange for improved runtime performance. Yet, existing data-driven methods fall short of the extreme memory and performance constraints imposed by modern interactive applications like AAA games and virtual reality. Here, performance budgets for physics simulation range from tens to hundreds of micro-seconds per frame, per object. We present a data-driven physical simulation method that meets these constraints. Our method combines subspace simulation techniques with machine learning which, when coupled, enables a very efficient subspace-only physics simulation that supports interactions with external objects – a longstanding challenge for existing sub-space techniques. We also present an interpretation of our method as a special case of subspace Verlet integration, where we apply machine learning to efficiently approximate the physical forces of the system directly in the subspace. We propose several practical solutions required to make effective use of such a model, including a novel training methodology required for prediction stability, and a GPU-friendly subspace decompression algorithm to accelerate rendering.
@inproceedings{DBLP:conf/sca/HoldenDDN19,
author = {Daniel Holden and
Bang Chi Duong and
Sayantan Datta and
Derek Nowrouzezahrai},
editor = {Sung{-}Hee Lee and
Craig A. Schroeder and
Stephen N. Spencer and
Christopher Batty and
Jin Huang},
title = {Subspace neural physics: fast data-driven interactive simulation},
booktitle = {Proceedings of the 18th annual {ACM} SIGGRAPH/Eurographics Symposium
on Computer Animation, {SCA} 2019, Los Angeles, CA, USA, July 26-28,
2019},
pages = {6:1--6:12},
publisher = {{ACM}},
year = {2019},
url = {https://doi.org/10.1145/3309486.3340245},
doi = {10.1145/3309486.3340245},
timestamp = {Fri, 12 Jun 2020 12:39:49 +0200},
biburl = {https://dblp.org/rec/conf/sca/HoldenDDN19.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Navigation Agents for the Visually Impaired: A Sidewalk Simulator and Experiments.
Weiss, M.; Chamorro, S.; Girgis, R.; Luck, M.; Kahou, S. E.; Cohen, J. P.; Nowrouzezahrai, D.; Precup, D.; Golemo, F.; and Pal, C.
CoRR, abs/1910.13249. 2019.
Paper
link
bibtex
abstract Millions of blind and visually-impaired (BVI) people navigate urban environments every day, using smartphones for high-level path-planning and white canes or guide dogs for local information. However, many BVI people still struggle to travel to new places. In our endeavor to create a navigation assistant for the BVI, we found that existing Reinforcement Learning (RL) environments were unsuitable for the task. This work introduces SEVN, a sidewalk simulation environment and a neural network-based approach to creating a navigation agent. SEVN contains panoramic images with labels for house numbers, doors, and street name signs, and formulations for several navigation tasks. We study the performance of an RL algorithm (PPO) in this setting. Our policy model fuses multi-modal observations in the form of variable resolution images, visible text, and simulated GPS data to navigate to a goal door. We hope that this dataset, simulator, and experimental results will provide a foundation for further research into the creation of agents that can assist members of the BVI community with outdoor navigation.
@article{DBLP:journals/corr/abs-1910-13249,
author = {Martin Weiss and
Simon Chamorro and
Roger Girgis and
Margaux Luck and
Samira Ebrahimi Kahou and
Joseph Paul Cohen and
Derek Nowrouzezahrai and
Doina Precup and
Florian Golemo and
Chris Pal},
title = {Navigation Agents for the Visually Impaired: {A} Sidewalk Simulator
and Experiments},
journal = {CoRR},
volume = {abs/1910.13249},
year = {2019},
url = {http://arxiv.org/abs/1910.13249},
eprinttype = {arXiv},
eprint = {1910.13249},
timestamp = {Thu, 31 Oct 2019 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1910-13249.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Robo-PlaNet: Learning to Poke in a Day.
Chevalier-Boisvert, M.; Alain, G.; Golemo, F.; and Nowrouzezahrai, D.
CoRR, abs/1911.03594. 2019.
Paper
link
bibtex
abstract Recently, the Deep Planning Network (PlaNet) approach was introduced as a model-based reinforcement learning method that learns environment dynamics directly from pixel observations. This architecture is useful for learning tasks in which either the agent does not have access to meaningful states (like position/velocity of robotic joints) or where the observed states significantly deviate from the physical state of the agent (which is commonly the case in low-cost robots in the form of backlash or noisy joint readings). PlaNet, by design, interleaves phases of training the dynamics model with phases of collecting more data on the target environment, leading to long training times. In this work, we introduce Robo-PlaNet, an asynchronous version of PlaNet. This algorithm consistently reaches higher performance in the same amount of time, which we demonstrate in both a simulated and a real robotic experiment.
@article{DBLP:journals/corr/abs-1911-03594,
author = {Maxime Chevalier{-}Boisvert and
Guillaume Alain and
Florian Golemo and
Derek Nowrouzezahrai},
title = {Robo-PlaNet: Learning to Poke in a Day},
journal = {CoRR},
volume = {abs/1911.03594},
year = {2019},
url = {http://arxiv.org/abs/1911.03594},
eprinttype = {arXiv},
eprint = {1911.03594},
timestamp = {Sun, 01 Dec 2019 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1911-03594.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2018
(7)
|
Integrating Clipped Spherical Harmonics Expansions.
Belcour, L.; Xie, G.; Hery, C.; Meyer, M.; Jarosz, W.; and Nowrouzezahrai, D.
ACM Trans. Graph., 37(2): 19. 2018.
Presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract Many applications in rendering rely on integrating functions over spherical polygons. We present a new numerical solution for computing the integral of spherical harmonics (SH) expansions clipped to polygonal domains. Our solution, based on zonal decompositions of spherical integrands and discrete contour integration, introduces an important numerical operating for SH expansions in rendering applications. Our method is simple, efficient, and scales linearly in the bandlimited integrand’s harmonic expansion. We apply our technique to problems in rendering, including surface and volume shading, hierarchical product importance sampling, and fast basis projection for interactive rendering. Moreover, we show how to handle general, nonpolynomial integrands in a Monte Carlo setting using control variates. Our technique computes the integral of bandlimited spherical functions with performance competitive to (or faster than) more general numerical integration methods for a broad class of problems, both in offline and interactive rendering contexts. Our implementation is simple, relying only on self-contained SH evaluation and discrete contour integration routines, and we release a full source CPU-only and shader-based implementations (<750 lines of commented code).
@article{DBLP:journals/tog/BelcourXHMJN18,
author = {Laurent Belcour and
Guofu Xie and
Christophe Hery and
Mark Meyer and
Wojciech Jarosz and
Derek Nowrouzezahrai},
title = {Integrating Clipped Spherical Harmonics Expansions},
journal = {{ACM} Trans. Graph.},
volume = {37},
number = {2},
pages = {19},
year = {2018},
url = {https://doi.org/10.1145/3015459},
doi = {10.1145/3015459},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/BelcourXHMJN18.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Gradient-domain volumetric photon density estimation.
Gruson, A.; Hua, B.; Vibert, N.; Nowrouzezahrai, D.; and Hachisuka, T.
ACM Trans. Graph., 37(4): 82. 2018.
Paper (PDF)
Paper
doi
link
bibtex
abstract Gradient-domain rendering can improve the convergence of surface-based light transport by exploiting smoothness in image space. Scenes with participating media exhibit similar smoothness and could potentially benefit from gradient-domain techniques. We introduce the first gradient-domain formulation of image synthesis with homogeneous participating media, including four novel and efficient gradient-domain volumetric density estimation algorithms. We show that naive extensions of gradient domain path-space and density estimation methods to volumetric media, while functional, can result in inefficient estimators. Focussing on point-, beam- and plane-based gradient-domain estimators, we introduce a novel shift mapping that eliminates redundancies in the naive formulations using spatial relaxation within the volume. We show that gradient-domain volumetric rendering improve convergence compared to primal domain state-of-the-art, across a suite of scenes. Our formulation and algorithms support progressive estimation and are easy to incorporate atop existing renderers.
@article{DBLP:journals/tog/GrusonHVNH18,
author = {Adrien Gruson and
Binh{-}Son Hua and
Nicolas Vibert and
Derek Nowrouzezahrai and
Toshiya Hachisuka},
title = {Gradient-domain volumetric photon density estimation},
journal = {{ACM} Trans. Graph.},
volume = {37},
number = {4},
pages = {82},
year = {2018},
url = {https://doi.org/10.1145/3197517.3201363},
doi = {10.1145/3197517.3201363},
timestamp = {Fri, 10 Jun 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/GrusonHVNH18.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Scalable appearance filtering for complex lighting effects.
Gamboa, L. E.; Guertin, J.; and Nowrouzezahrai, D.
ACM Trans. Graph., 37(6): 277. 2018.
Paper (PDF)
Paper
doi
link
bibtex
2 downloads
abstract Realistic rendering with materials that exhibit high-frequency spatial variation remains a challenge, as eliminating spatial and temporal aliasing requires prohibitively high sampling rates. Recent work has made the problem more tractable, however existing methods remain prohibitively expensive when using large environmental lights and/or (correctly filtered) global illumination. We present an appearance model with explicit high-frequency micro-normal variation, and a filtering approach that scales to multi-dimensional shading integrals. By combining a novel and compact half-vector histogram scheme with a directional basis expansion, we accurately compute the integral of filtered high-frequency reflectance over large lights with angularly varying emission. Our approach is scalable, rendering images indistinguishable from ground truth at over 10x the speed of the state-of-the-art and with only 15% the memory footprint. When filtering appearance with global illumination, we outperform the state-of-the-art by ~30x.
@article{DBLP:journals/tog/GamboaGN18,
author = {Luis E. Gamboa and
Jean{-}Philippe Guertin and
Derek Nowrouzezahrai},
title = {Scalable appearance filtering for complex lighting effects},
journal = {{ACM} Trans. Graph.},
volume = {37},
number = {6},
pages = {277},
year = {2018},
url = {https://doi.org/10.1145/3272127.3275058},
doi = {10.1145/3272127.3275058},
timestamp = {Fri, 10 Jun 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/GamboaGN18.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Matrix Bidirectional Path Tracing.
Chaitanya, C. R. A.; Belcour, L.; Hachisuka, T.; Premoze, S.; Pantaleoni, J.; and Nowrouzezahrai, D.
In Jakob, W.; and Hachisuka, T., editor(s), 29th Eurographics Symposium on Rendering, Experimental Ideas & Implementations, EGSR 2018, EI&I Track, Karlsruhe, Germany, 1-4 July 2018, pages 23–32, 2018. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract Sampled paths in Monte Carlo ray tracing can be arbitrarily close to each other due to its stochastic nature. Such clumped samples in the path space tend to contribute little toward an accurate estimate of each pixel. Bidirectional light transport methods make this issue further complicated since connecting paths of sampled subpaths can be arbitrarily clumped again. We propose a matrix formulation of bidirectional light transport that enables stratification (and low-discrepancy sampling) in this connection space. This stratification allows us to distribute computation evenly across contributing paths in the image, which is not possible with standard bidirectional or Markov chain solutions. Each element in our matrix formulation represents a pair of connected eye- and light-subpaths. By carefully reordering these elements, we build a 2D space where equally contributing paths are distributed coherently. We devise an unbiased rendering algorithm that leverages this coherence to effectively sample path space, consistently achieving a 2-3x speedup in radiometrically complex scenes compared to the state-of-the-art.
@inproceedings{DBLP:conf/rt/ChaitanyaBHPPN18,
author = {Chakravarty Reddy Alla Chaitanya and
Laurent Belcour and
Toshiya Hachisuka and
Simon Premoze and
Jacopo Pantaleoni and
Derek Nowrouzezahrai},
editor = {Wenzel Jakob and
Toshiya Hachisuka},
title = {Matrix Bidirectional Path Tracing},
booktitle = {29th Eurographics Symposium on Rendering, Experimental Ideas {\&}
Implementations, {EGSR} 2018, EI{\&}I Track, Karlsruhe, Germany, 1-4
July 2018},
pages = {23--32},
publisher = {Eurographics Association},
year = {2018},
url = {https://doi.org/10.2312/sre.20181169},
doi = {10.2312/SRE.20181169},
timestamp = {Tue, 29 Jan 2019 17:35:36 +0100},
biburl = {https://dblp.org/rec/conf/rt/ChaitanyaBHPPN18.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Proceedings of the ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, I3D 2018, Montreal, QC, Canada, May 15-18, 2018.
Spencer, S. N.; McGuire, M.; and Nowrouzezahrai, D.,
editors.
ACM. 2018.
General Program Co-chair
Paper
doi
link
bibtex
abstract This was the 22nd annual event in the series of very successful ACM/SIGGRAPH Symposia on Interactive 3D Graphics and Games. The conference took place May 15-18, 2018 at the Unity Studios in Montreal, Canada.
@proceedings{DBLP:conf/si3d/2018,
editor = {Stephen N. Spencer and
Morgan McGuire and
Derek Nowrouzezahrai},
title = {Proceedings of the {ACM} {SIGGRAPH} Symposium on Interactive 3D Graphics
and Games, {I3D} 2018, Montreal, QC, Canada, May 15-18, 2018},
publisher = {{ACM}},
year = {2018},
url = {https://doi.org/10.1145/3190834},
doi = {10.1145/3190834},
timestamp = {Wed, 21 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/si3d/2018.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Adversarial Geometry and Lighting using a Differentiable Renderer.
Liu, H. D.; Tao, M.; Li, C.; Nowrouzezahrai, D.; and Jacobson, A.
CoRR, abs/1808.02651. 2018.
Paper
link
bibtex
@article{DBLP:journals/corr/abs-1808-02651,
author = {Hsueh{-}Ti Derek Liu and
Michael Tao and
Chun{-}Liang Li and
Derek Nowrouzezahrai and
Alec Jacobson},
title = {Adversarial Geometry and Lighting using a Differentiable Renderer},
journal = {CoRR},
volume = {abs/1808.02651},
year = {2018},
url = {http://arxiv.org/abs/1808.02651},
eprinttype = {arXiv},
eprint = {1808.02651},
timestamp = {Mon, 24 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-1808-02651.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2017
(9)
|
Gradient-Domain Photon Density Estimation.
Hua, B.; Gruson, A.; Nowrouzezahrai, D.; and Hachisuka, T.
Comput. Graph. Forum, 36(2): 31–38. 2017.
Paper (PDF) Slides (PPTx)
Paper
doi
link
bibtex
abstract The most common solutions to the light transport problem rely on either Monte Carlo (MC) integration or density estimation methods, such as uni- & bi-directional path tracing or photon mapping. Recent gradient-domain extensions of MC approaches show great promise; here, gradients of the final image are estimated numerically (instead of the image intensities themselves) with coherent paths generated from a deterministic shift mapping. We extend gradient-domain approaches to light transport simulation based on density estimation. As with previous gradient-domain methods, we detail important considerations that arise when moving from a primal- to gradient-domain estimator. We provide an efficient and straightforward solution to these problems. Our solution supports stochastic progressive density estimation, so it is robust to complex transport effects. We show that gradient-domain photon density estimation converges faster than its primal-domain counterpart, as well as being generally more robust than gradient-domain uni- & bi-directional path tracing for scenes dominated by complex transport.
@article{DBLP:journals/cgf/HuaGNH17,
author = {Binh{-}Son Hua and
Adrien Gruson and
Derek Nowrouzezahrai and
Toshiya Hachisuka},
title = {Gradient-Domain Photon Density Estimation},
journal = {Comput. Graph. Forum},
volume = {36},
number = {2},
pages = {31--38},
year = {2017},
url = {https://doi.org/10.1111/cgf.13104},
doi = {10.1111/CGF.13104},
timestamp = {Fri, 16 Jun 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/HuaGNH17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Antialiasing Complex Global Illumination Effects in Path-Space.
Belcour, L.; Yan, L.; Ramamoorthi, R.; and Nowrouzezahrai, D.
ACM Trans. Graph., 36(1): 9:1–9:13. 2017.
Presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
abstract We present the first method to efficiently predict antialiasing footprints to pre-filter color-, normal-, and displacement-mapped appearance in the context of multi-bounce global illumination. We derive Fourier spectra for radiance and importance functions that allow us to compute spatial-angular filtering footprints at path vertices, for both uni- and bi-directional path construction. We then use these footprints to antialias reflectance modulated by high-resolution maps (such as color and normal maps) encountered along a path. In doing so, we also unify the traditional path-space formulation of light-transport with our frequency-space interpretation of global illumination pre-filtering. Our method is fully compatible with all existing single bounce pre-filtering appearance models, not restricted by path length, and easy to implement atop existing path-space renderers. We illustrate its effectiveness on several radiometrically complex scenarios where previous approaches either completely fail or require orders of magnitude more time to arrive at similarly high-quality results.
@article{DBLP:journals/tog/BelcourYRN17,
author = {Laurent Belcour and
Ling{-}Qi Yan and
Ravi Ramamoorthi and
Derek Nowrouzezahrai},
title = {Antialiasing Complex Global Illumination Effects in Path-Space},
journal = {{ACM} Trans. Graph.},
volume = {36},
number = {1},
pages = {9:1--9:13},
year = {2017},
url = {https://doi.org/10.1145/2990495},
doi = {10.1145/2990495},
timestamp = {Mon, 21 Aug 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/BelcourYRN17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Interactive reconstruction of Monte Carlo image sequences using a recurrent denoising autoencoder.
Chaitanya, C. R. A.; Kaplanyan, A. S.; Schied, C.; Salvi, M.; Lefohn, A. E.; Nowrouzezahrai, D.; and Aila, T.
ACM Trans. Graph., 36(4): 98:1–98:12. 2017.
Paper (PDF)
Paper
doi
link
bibtex
4 downloads
abstract We describe a machine learning technique for reconstructing image se- quences rendered using Monte Carlo methods. Our primary focus is on reconstruction of global illumination with extremely low sampling budgets at interactive rates. Motivated by recent advances in image restoration with deep convolutional networks, we propose a variant of these networks better suited to the class of noise present in Monte Carlo rendering. We allow for much larger pixel neighborhoods to be taken into account, while also improving execution speed by an order of magnitude. Our primary contri- bution is the addition of recurrent connections to the network in order to drastically improve temporal stability for sequences of sparsely sampled input images. Our method also has the desirable property of automatically modeling relationships based on auxiliary per-pixel input channels, such as depth and normals. We show signi cantly higher quality results compared to existing methods that run at comparable speeds, and furthermore argue a clear path for making our method run at realtime rates in the near future.
@article{DBLP:journals/tog/ChaitanyaKSSLNA17,
author = {Chakravarty R. Alla Chaitanya and
Anton S. Kaplanyan and
Christoph Schied and
Marco Salvi and
Aaron E. Lefohn and
Derek Nowrouzezahrai and
Timo Aila},
title = {Interactive reconstruction of Monte Carlo image sequences using a
recurrent denoising autoencoder},
journal = {{ACM} Trans. Graph.},
volume = {36},
number = {4},
pages = {98:1--98:12},
year = {2017},
url = {https://doi.org/10.1145/3072959.3073601},
doi = {10.1145/3072959.3073601},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/ChaitanyaKSSLNA17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Ballistic Shadow Art.
Chen, X.; Andrews, S.; Nowrouzezahrai, D.; and Kry, P. G.
In Eisemann, E.; and Bateman, S., editor(s), Proceedings of the 43rd Graphics Interface Conference 2017, Edmonton, Alberta, Canada, May 16-19, 2017, pages 190–198, 2017. Canadian Human-Computer Communications Society / ACM
Paper (PDF) Slides (PPTx)
Paper
doi
link
bibtex
1 download
abstract We present a framework for generating animated shadow art using occluders under ballistic motion. We apply a stochastic optimization to find the parameters of a multi-body physics simulation that produce a desired shadow at a specific instant in time. We perform simulations across many different initial conditions, applying a set of carefully crafted energy functions to evaluate the motion trajectory and multi-body shadows. We select the optimal parameters, resulting in a ballistics simulation that produces ephemeral shadow art. Users can design physically-plausible dynamic artwork that would be extremely challenging if even possible to achieve manually. We present and analyze a number of compelling examples.
@inproceedings{DBLP:conf/graphicsinterface/ChenANK17,
author = {Xiaozhong Chen and
Sheldon Andrews and
Derek Nowrouzezahrai and
Paul G. Kry},
editor = {Elmar Eisemann and
Scott Bateman},
title = {Ballistic Shadow Art},
booktitle = {Proceedings of the 43rd Graphics Interface Conference 2017, Edmonton,
Alberta, Canada, May 16-19, 2017},
pages = {190--198},
publisher = {Canadian Human-Computer Communications Society / {ACM}},
year = {2017},
url = {https://doi.org/10.20380/GI2017.24},
doi = {10.20380/GI2017.24},
timestamp = {Fri, 04 Oct 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/graphicsinterface/ChenANK17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Frequency Based Radiance Cache for Rendering Animations.
Dubouchet, R. A.; Belcour, L.; and Nowrouzezahrai, D.
In Zwicker, M.; and Sander, P. V., editor(s), 28th Eurographics Symposium on Rendering, Rendering - Experimental Ideas & Implementations, EGSR 2017, EI&I Track, Helsinki, Finland, 19-21 June 2017, pages 41–53, 2017. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract We propose a method to render animation sequences with direct distant lighting that only shades a fraction of the total pixels. We leverage frequency-based analyses of light transport to determine shading and image sampling rates across an animation using a samples cache. To do so, we derive frequency bandwidths that account for the complexity of distant lights, visibility, BRDF, and temporal coherence during animation. We finaly apply a cross-bilateral filter when rendering our final images from sparse sets of shading points placed according to our frequency-based oracles (generally less than 25% of the pixels, per frame).
@inproceedings{DBLP:conf/rt/DubouchetBN17,
author = {Renaud Adrien Dubouchet and
Laurent Belcour and
Derek Nowrouzezahrai},
editor = {Matthias Zwicker and
Pedro V. Sander},
title = {Frequency Based Radiance Cache for Rendering Animations},
booktitle = {28th Eurographics Symposium on Rendering, Rendering - Experimental
Ideas {\&} Implementations, {EGSR} 2017, EI{\&}I Track, Helsinki,
Finland, 19-21 June 2017},
pages = {41--53},
publisher = {Eurographics Association},
year = {2017},
url = {https://doi.org/10.2312/sre.20171193},
doi = {10.2312/SRE.20171193},
timestamp = {Wed, 09 Jan 2019 07:20:11 +0100},
biburl = {https://dblp.org/rec/conf/rt/DubouchetBN17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Gradient-Domain Vertex Connection and Merging.
Sun, W.; Sun, X.; Carr, N. A.; Nowrouzezahrai, D.; and Ramamoorthi, R.
In Zwicker, M.; and Sander, P. V., editor(s), 28th Eurographics Symposium on Rendering, Rendering - Experimental Ideas & Implementations, EGSR 2017, EI&I Track, Helsinki, Finland, 19-21 June 2017, pages 83–92, 2017. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract Recently, gradient-domain rendering techniques have shown great promise in reducing Monte Carlo noise and improving over- all rendering efficiency. However, all existing gradient-domain methods are built exclusively on top of Monte Carlo integration or density estimation. While these methods can be effective, combining Monte Carlo integration and density estimation has been shown (in the primal domain) to more robustly handle a wider variety of light paths from arbitrarily complex scenes. We present gradient-domain vertex connection and merging (G-VCM), a new gradient-domain technique motivated by primal domain VCM. Our method enables robust gradient sampling in the presence of complex transport, such as specular-diffuse-specular paths, while retaining the denoising power and fast convergence of gradient-domain bidirectional path tracing. We show that G-VCM is able to handle a variety of scenes that exhibit slow convergence when rendered with previous gradient-domain methods.
@inproceedings{DBLP:conf/rt/Sun0CNR17,
author = {Weilun Sun and
Xin Sun and
Nathan A. Carr and
Derek Nowrouzezahrai and
Ravi Ramamoorthi},
editor = {Matthias Zwicker and
Pedro V. Sander},
title = {Gradient-Domain Vertex Connection and Merging},
booktitle = {28th Eurographics Symposium on Rendering, Rendering - Experimental
Ideas {\&} Implementations, {EGSR} 2017, EI{\&}I Track, Helsinki,
Finland, 19-21 June 2017},
pages = {83--92},
publisher = {Eurographics Association},
year = {2017},
url = {https://doi.org/10.2312/sre.20171197},
doi = {10.2312/SRE.20171197},
timestamp = {Tue, 07 Sep 2021 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/rt/Sun0CNR17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Real-time global illumination using precomputed light field probes.
McGuire, M.; Mara, M.; Nowrouzezahrai, D.; and Luebke, D.
In Spencer, S. N.; Bargteil, A. W.; and Mitchell, K., editor(s), Proceedings of the 21st ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games, I3D 2017, San Francisco, CA, USA, March 4-5, 2017, pages 2:1–2:11, 2017. ACM
Best Presentation Award Winner!
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract We introduce a new data structure and algorithms that employ it to compute real-time global illumination from static environments. Light field probes encode a scene's full light field and internal visibility. They extend current radiance and irradiance probe structures with per-texel visibility information similar to a G-buffer and variance shadow map. We apply ideas from screen-space and voxel cone tracing techniques to this data structure to efficiently sample radiance on world space rays, with correct visibility information, directly within pixel and compute shaders. From these primitives, we then design two GPU algorithms to efficiently gather real-time, viewer-dependent global illumination onto both static and dynamic objects. These algorithms make different tradeoffs between performance and accuracy. Supplemental GLSL source code is included.
@inproceedings{DBLP:conf/si3d/McGuireMNL17,
author = {Morgan McGuire and
Mike Mara and
Derek Nowrouzezahrai and
David Luebke},
editor = {Stephen N. Spencer and
Adam W. Bargteil and
Kenny Mitchell},
title = {Real-time global illumination using precomputed light field probes},
booktitle = {Proceedings of the 21st {ACM} {SIGGRAPH} Symposium on Interactive
3D Graphics and Games, {I3D} 2017, San Francisco, CA, USA, March 4-5,
2017},
pages = {2:1--2:11},
publisher = {{ACM}},
year = {2017},
url = {https://doi.org/10.1145/3023368.3023378},
doi = {10.1145/3023368.3023378},
timestamp = {Tue, 06 Nov 2018 16:59:21 +0100},
biburl = {https://dblp.org/rec/conf/si3d/McGuireMNL17.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2016
(4)
|
State of the Art in Artistic Editing of Appearance, Lighting and Material.
Schmidt, T.; Pellacini, F.; Nowrouzezahrai, D.; Jarosz, W.; and Dachsbacher, C.
Comput. Graph. Forum, 35(1): 216–233. 2016.
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract Mimicking the appearance of the real world is a longstanding goal of computer graphics, with several important applications in the feature-film, architecture and medical industries. Images with well-designed shading are an important tool for conveying information about the world, be it the shape and function of a CAD model, or the mood of a movie sequence. However, authoring this content is often a tedious task, even if undertaken by groups of highly-trained and experienced artists. Unsurprisingly, numerous methods to facilitate and accelerate this appearance editing task have been proposed, enabling the editing of scene objects' appearances, lighting, and materials, as well as entailing the introduction of new interaction paradigms and specialized preview rendering techniques. In this review we provide a comprehensive survey of artistic appearance, lighting, and material editing approaches. We organize this complex and active research area in a structure tailored to academic researchers, graduate students, and industry professionals alike. In addition to editing approaches, we discuss how user interaction paradigms and rendering backends combine to form usable systems for appearance editing. We conclude with a discussion of open problems and challenges to motivate and guide future research.
@article{DBLP:journals/cgf/SchmidtPNJD16,
author = {Thorsten{-}Walther Schmidt and
Fabio Pellacini and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Carsten Dachsbacher},
title = {State of the Art in Artistic Editing of Appearance, Lighting and Material},
journal = {Comput. Graph. Forum},
volume = {35},
number = {1},
pages = {216--233},
year = {2016},
url = {https://doi.org/10.1111/cgf.12721},
doi = {10.1111/CGF.12721},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/SchmidtPNJD16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Reduced Aggregate Scattering Operators for Path Tracing.
Blumer, A.; Novák, J.; Habel, R.; Nowrouzezahrai, D.; and Jarosz, W.
Comput. Graph. Forum, 35(7): 461–473. 2016.
Paper (PDF) Slides (PPTx)
Paper
doi
link
bibtex
abstract Aggregate scattering operators (ASOs) describe the overall scattering behavior of an asset (i.e., an object or volume, or collection thereof) accounting for all orders of its internal scattering. We propose a practical way to precompute and compactly store ASOs and demonstrate their ability to accelerate path tracing. Our approach is modular avoiding costly and inflexible scene-dependent precomputation. This is achieved by decoupling light transport within and outside of each asset, and precomputing on a per-asset level. We store the internal transport in a reduced-dimensional subspace tailored to the structure of the asset geometry, its scattering behavior, and typical illumination conditions, allowing the ASOs to maintain good accuracy with modest memory requirements. The precomputed ASO can be reused across all instances of the asset and across multiple scenes. We augment ASOs with functionality enabling multi-bounce importance sampling, fast short-circuiting of complex light paths, and compact caching, while retaining rapid progressive preview rendering. We demonstrate the benefits of our ASOs by efficiently path tracing scenes containing many instances of objects with complex inter-reflections or multiple scattering.
@article{DBLP:journals/cgf/BlumerNHNJ16,
author = {Adrian Blumer and
Jan Nov{\'{a}}k and
Ralf Habel and
Derek Nowrouzezahrai and
Wojciech Jarosz},
title = {Reduced Aggregate Scattering Operators for Path Tracing},
journal = {Comput. Graph. Forum},
volume = {35},
number = {7},
pages = {461--473},
year = {2016},
url = {https://doi.org/10.1111/cgf.13043},
doi = {10.1111/CGF.13043},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/BlumerNHNJ16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Non-Parametric Factor Microfacet Model for Isotropic BRDFs.
Bagher, M. M.; Snyder, J. M.; and Nowrouzezahrai, D.
ACM Trans. Graph., 35(5): 159:1–159:16. 2016.
To be presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
2 downloads
abstract We investigate the expressiveness of the microfacet model for isotropic BRDFs measured from real materials by introducing a non-parametric factor model that represents the model's functional structure but abandons restricted parametric formulations of its factors. We propose a new objective based on compressive weighting that controls rendering error in high dynamic range BRDF fits better than previous factorization approaches. We develop a simple numerical procedure to minimize this objective and handle dependencies that arise between microfacet factors. Our method faithfully captures a more comprehensive set of materials than previous state-of-the-art parametric approaches, yet remains compact (3.2KB per BRDF). We experimentally validate the benefit of the microfacet model over a naive orthogonal factorization, and show that fidelity for diffuse materials is modestly improved by fitting an unrestricted shadowing/masking factor. We also compare against a recent data-driven factorization approach [Bilgili et al. 2011] and show that our microfacet-based representation improves rendering accuracy for most materials while reducing storage by more than 10x.
@article{DBLP:journals/tog/BagherSN16,
author = {Mahdi M. Bagher and
John M. Snyder and
Derek Nowrouzezahrai},
title = {A Non-Parametric Factor Microfacet Model for Isotropic BRDFs},
journal = {{ACM} Trans. Graph.},
volume = {35},
number = {5},
pages = {159:1--159:16},
year = {2016},
url = {https://doi.org/10.1145/2907941},
doi = {10.1145/2907941},
timestamp = {Mon, 13 May 2019 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/BagherSN16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Deep g-buffers for stable global illumination approximation.
Mara, M.; McGuire, M.; Nowrouzezahrai, D.; and Luebke, D. P.
In Luebke, D.; and Molnar, S., editor(s), Proceedings of High Performance Graphics, HPG 2016, Dublin, Ireland, June 20-22, 2016, pages 87–98, 2016. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract We introduce a new hardware-accelerated method for constructing Deep G-buffers that is 2x-8x faster than the previous depth peeling method and produces more stable results. We then build several high-performance shading algorithms atop our representation, including dynamic diffuse interreflection, ambient occlusion (AO), and mirror reflection effects. Our construction method s order-independent, guarantees a minimum separation between layers, operates in a (small) bounded memory footprint, and does not require per-pixel sorting. Moreover, addressing the increasingly expensive cost of pre-rasterization, our approach requires only a single pass over the scene geometry. Our global illumination algorithms approach the speed of the fastest screen-space AO-only techniques while significantly exceeding their quality: we capture small-scale details and complex radiometric effects more robustly than screen-space techniques, and we implicitly handle dynamic illumination conditions. We include the pseudocode for our Deep G-buffer construction in the paper and the full source code of our technique in our supplemental document.
@inproceedings{DBLP:conf/egh/MaraMNL16,
author = {Michael Mara and
Morgan McGuire and
Derek Nowrouzezahrai and
David P. Luebke},
editor = {David Luebke and
Steven Molnar},
title = {Deep g-buffers for stable global illumination approximation},
booktitle = {Proceedings of High Performance Graphics, {HPG} 2016, Dublin, Ireland,
June 20-22, 2016},
pages = {87--98},
publisher = {Eurographics Association},
year = {2016},
url = {https://doi.org/10.2312/hpg.20161195},
doi = {10.2312/HPG.20161195},
timestamp = {Wed, 24 May 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/egh/MaraMNL16.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2015
(6)
|
Forward to the special section on SIBGRAPI 2014.
Nowrouzezahrai, D.; and Nehab, D.
Comput. Graph., 53: A3. 2015.
Paper
doi
link
bibtex
@article{DBLP:journals/cg/NowrouzezahraiN15,
author = {Derek Nowrouzezahrai and
Diego Nehab},
title = {Forward to the special section on {SIBGRAPI} 2014},
journal = {Comput. Graph.},
volume = {53},
pages = {A3},
year = {2015},
url = {https://doi.org/10.1016/j.cag.2015.10.008},
doi = {10.1016/J.CAG.2015.10.008},
timestamp = {Wed, 19 Feb 2020 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/cg/NowrouzezahraiN15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Efficient and Accurate Spherical Kernel Integrals Using Isotropic Decomposition.
Soler, C.; Bagher, M. M.; and Nowrouzezahrai, D.
ACM Trans. Graph., 34(5): 161:1–161:14. 2015.
Presented at SIGGRAPH Asia
Paper (PDF)
Paper
doi
link
bibtex
abstract Spherical filtering is fundamental to many problems in image synthesis, such as computing the reflected light over a surface or anti-aliasing mirror reflections over a pixel. This operation is challenging since the profile of spherical filters (e.g., the view-evaluated BRDF or the geometry-warped pixel footprint, above) typically exhibits both spatial- and rotational variation at each pixel, precluding precomputed solutions. We accelerate complex spherical filtering tasks using isotropic spherical decomposition (ISD), decomposing spherical filters into a linear combination of simpler isotropic kernels. Our general ISD is flexible to the choice of the isotropic kernels, and we demonstrate practical realizations of ISD on several problems in rendering: shading and prefiltering with spatially-varying BRDFs, anti-aliasing environment mapped mirror reflections, and filtering of noisy reflectance data. Compared to previous basis-space rendering solutions, our shading solution generates ground truth-quality results at interactive rates, avoiding costly reconstruction and large approximation errors.
@article{DBLP:journals/tog/SolerBN15,
author = {Cyril Soler and
Mahdi M. Bagher and
Derek Nowrouzezahrai},
title = {Efficient and Accurate Spherical Kernel Integrals Using Isotropic
Decomposition},
journal = {{ACM} Trans. Graph.},
volume = {34},
number = {5},
pages = {161:1--161:14},
year = {2015},
url = {https://doi.org/10.1145/2797136},
doi = {10.1145/2797136},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/SolerBN15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Surface turbulence for particle-based liquid simulations.
Mercier, O.; Beauchemin, C.; Thuerey, N.; Kim, T.; and Nowrouzezahrai, D.
ACM Trans. Graph., 34(6): 202:1–202:10. 2015.
Paper (PDF)
Paper
doi
link
bibtex
abstract We present a method to increase the apparent resolution of particlebased liquid simulations. Our method first outputs a dense, temporally coherent, regularized point set from a coarse particle-based liquid simulation. We then apply a surface-only Lagrangian wave simulation to this high-resolution point set. We develop novel methods for seeding and simulating waves over surface points, and use them to generate high-resolution details. We avoid error-prone surface mesh processing, and robustly propagate waves without the need for explicit connectivity information. Our seeding strategy combines a robust curvature evaluation with multiple bands of seeding oscillators, injects waves with arbitrarily fine-scale structures, and properly handles obstacle boundaries. We generate detailed fluid surfaces from coarse simulations as an independent post-process that can be applied to most particle-based fluid solvers.
@article{DBLP:journals/tog/MercierBTKN15,
author = {Olivier Mercier and
Cynthia Beauchemin and
Nils Thuerey and
Theodore Kim and
Derek Nowrouzezahrai},
title = {Surface turbulence for particle-based liquid simulations},
journal = {{ACM} Trans. Graph.},
volume = {34},
number = {6},
pages = {202:1--202:10},
year = {2015},
url = {https://doi.org/10.1145/2816795.2818115},
doi = {10.1145/2816795.2818115},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/MercierBTKN15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Practical Shading of Height Fields and Meshes using Spherical Harmonic Exponentiation.
Giraud, A.; and Nowrouzezahrai, D.
In Lehtinen, J.; and Nowrouzezahrai, D., editor(s), 26th Eurographics Symposium on Rendering, Rendering - Experimental Ideas & Implementations, EGSR 2015, EI&I Track, Darmstadt, Germany, June 23-26, 2015, pages 1–8, 2015. Eurographics Association
Paper (PDF) Slides (PDF)
Paper
doi
link
bibtex
abstract Interactively computing smooth shading effects from environmental lighting, such as soft shadows and glossy reflections, is a challenge in scenes with dynamic objects. We present a method to efficiently approximate these effects in scenes comprising animating objects and dynamic height fields, additionally allowing interactive manipulation of the view and lighting. Our method extends spherical harmonic (SH) exponentiation approaches to support environmental shadowing from both dynamic blockers and dynamic height field geometry. We also derive analytic expressions for the view-evaluated BRDF, directly in the log-SH space, in order to support diffuse-to-glossy shadowed reflections while avoiding expensive basis-space product operations. We illustrate interactive rendering results using a hybrid, multi-resolution screen- and object-space visibility-marching algorithm that decouples geometric complexity from shading complexity.
@inproceedings{DBLP:conf/rt/GiraudN15,
author = {Aude Giraud and
Derek Nowrouzezahrai},
editor = {Jaakko Lehtinen and
Derek Nowrouzezahrai},
title = {Practical Shading of Height Fields and Meshes using Spherical Harmonic
Exponentiation},
booktitle = {26th Eurographics Symposium on Rendering, Rendering - Experimental
Ideas {\&} Implementations, {EGSR} 2015, EI{\&}I Track, Darmstadt,
Germany, June 23-26, 2015},
pages = {1--8},
publisher = {Eurographics Association},
year = {2015},
url = {https://doi.org/10.2312/sre.20151161},
doi = {10.2312/SRE.20151161},
timestamp = {Wed, 09 Jan 2019 07:30:05 +0100},
biburl = {https://dblp.org/rec/conf/rt/GiraudN15.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
26th Eurographics Symposium on Rendering, Rendering - Experimental Ideas & Implementations, EGSR 2015, EI&I Track, Darmstadt, Germany, June 23-26, 2015.
Lehtinen, J.; and Nowrouzezahrai, D.,
editors.
Eurographics Association. 2015.
Technical Papers Co-chair
Paper
link
bibtex
abstract This was the 26th annual event in the series of very successful Eurographics Symposia on Rendering and Eurographics Workshops on Rendering. The workshop took place June 23–26, 2015 at the Fraunhofer IGD, Darmstadt, Germany.
@proceedings{DBLP:conf/rt/2015eii,
editor = {Jaakko Lehtinen and
Derek Nowrouzezahrai},
title = {26th Eurographics Symposium on Rendering, Rendering - Experimental
Ideas {\&} Implementations, {EGSR} 2015, EI{\&}I Track, Darmstadt,
Germany, June 23-26, 2015},
publisher = {Eurographics Association},
year = {2015},
url = {https://diglib.eg.org/handle/10.2312/12636},
timestamp = {Wed, 09 Jan 2019 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/rt/2015eii.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2014
(8)
|
Visibility Silhouettes for Semi-Analytic Spherical Integration.
Nowrouzezahrai, D.; Baran, I.; Mitchell, K.; and Jarosz, W.
Comput. Graph. Forum, 33(1): 105–117. 2014.
Paper
doi
link
bibtex
4 downloads
abstract At each shade point, the spherical visibility function encodes occlusion from surrounding geometry, in all directions. Computing this function is difficult and point-sampling approaches, such as ray-tracing or hardware shadow mapping, are traditionally used to efficiently approximate it. We propose a semi-analytic solution to the problem where the spherical silhouette of the visibility is computed using a search over a 4D dual mesh of the scene. Once computed, we are able to semi-analytically integrate visibility-masked spherical functions along the visibility silhouette, instead of over the entire hemisphere. In this way, we avoid the artefacts that arise from using point-sampling strategies to integrate visibility, a function with unbounded frequency content. We demonstrate our approach on several applications, including direct illumination from realistic lighting and computation of pre-computed radiance transfer data. Additionally, we present a new frequency-space method for exactly computing all-frequency shadows on diffuse surfaces. Our results match ground truth computed using importance-sampled stratified Monte Carlo ray-tracing, with comparable performance on scenes with low-to-moderate geometric complexity.
@article{DBLP:journals/cgf/NowrouzezahraiBMJ14,
author = {Derek Nowrouzezahrai and
Ilya Baran and
Kenny Mitchell and
Wojciech Jarosz},
title = {Visibility Silhouettes for Semi-Analytic Spherical Integration},
journal = {Comput. Graph. Forum},
volume = {33},
number = {1},
pages = {105--117},
year = {2014},
url = {https://doi.org/10.1111/cgf.12257},
doi = {10.1111/CGF.12257},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/NowrouzezahraiBMJ14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Error analysis of estimators that use combinations of stochastic sampling strategies for direct illumination.
Subr, K.; Nowrouzezahrai, D.; Jarosz, W.; Kautz, J.; and Mitchell, K.
Comput. Graph. Forum, 33(4): 93–102. 2014.
Paper (PDF)
Paper
doi
link
bibtex
abstract We present a theoretical analysis of error of combinations of Monte Carlo estimators used in image synthesis. Importance sampling and multiple importance sampling are popular variance-reduction strategies. Unfortunately, neither strategy improves the rate of convergence of Monte Carlo integration. Jittered sampling (a type of stratified sampling), on the other hand is known to improve the convergence rate. Most rendering software optimistically combine importance sampling with jittered sampling, hoping to achieve both. We derive the exact error of the combination of multiple importance sampling with jittered sampling. In addition, we demonstrate a further benefit of introducing negative correlations (antithetic sampling) between estimates to the convergence rate. As with importance sampling, antithetic sampling is known to reduce error for certain classes of integrands without affecting the convergence rate. In this paper, our analysis and experiments reveal that importance and antithetic sampling, if used judiciously and in conjunction with jittered sampling, may improve convergence rates. We show the impact of such combinations of strategies on the convergence rate of estimators for direct illumination.
@article{DBLP:journals/cgf/SubrNJKM14,
author = {Kartic Subr and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Jan Kautz and
Kenny Mitchell},
title = {Error analysis of estimators that use combinations of stochastic sampling
strategies for direct illumination},
journal = {Comput. Graph. Forum},
volume = {33},
number = {4},
pages = {93--102},
year = {2014},
url = {https://doi.org/10.1111/cgf.12416},
doi = {10.1111/CGF.12416},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/SubrNJKM14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Unifying points, beams, and paths in volumetric light transport simulation.
Krivánek, J.; Georgiev, I.; Hachisuka, T.; Vévoda, P.; Sik, M.; Nowrouzezahrai, D.; and Jarosz, W.
ACM Trans. Graph., 33(4): 103:1–103:13. 2014.
Winner of the CGF Cover Image Competition!
Paper (PDF)
Paper
doi
link
bibtex
abstract Efficiently computing light transport in participating media in a manner that is robust to variations in media density, scattering albedo, and anisotropy is a difficult and important problem in realistic image synthesis. While many specialized rendering techniques can efficiently resolve subsets of transport in specific volumetric media, no single approach can robustly handle all types of effects. To combat this problem we unify volumetric density estimation, using point- and beam-estimates, and Monte Carlo-based solutions to the path-integral formulation of the rendering and radiative transport equations. We generalize multiple importance sampling to correctly handle combinations of these fundamentally different classes of radiance estimators. This in turn allows us to develop a single rendering algorithm that correctly combines the benefits (and mediates the limitations) of these powerful volume rendering techniques.
@article{DBLP:journals/tog/KrivanekGHVSNJ14,
author = {Jaroslav Kriv{\'{a}}nek and
Iliyan Georgiev and
Toshiya Hachisuka and
Petr V{\'{e}}voda and
Martin Sik and
Derek Nowrouzezahrai and
Wojciech Jarosz},
title = {Unifying points, beams, and paths in volumetric light transport simulation},
journal = {{ACM} Trans. Graph.},
volume = {33},
number = {4},
pages = {103:1--103:13},
year = {2014},
url = {https://doi.org/10.1145/2601097.2601219},
doi = {10.1145/2601097.2601219},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/KrivanekGHVSNJ14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Hierarchical diffusion curves for accurate automatic image vectorization.
Xie, G.; Sun, X.; Tong, X.; and Nowrouzezahrai, D.
ACM Trans. Graph., 33(6): 230:1–230:11. 2014.
Paper (PDF)
Paper
doi
link
bibtex
abstract Diffusion curve primitives are a compact and powerful representation for vector images. While several vector image authoring tools leverage these representations, automatically and accurately vectorizing arbitrary raster images using diffusion curves remains a difficult problem. We automatically generate sparse diffusion curve vectorizations of raster images by fitting curves in the Laplacian domain. Our approach is fast, combines Laplacian and bilaplacian diffusion curve representations, and generates a hierarchical representation that accurately reconstructs both vector art and natural images. The key idea of our method is to trace curves in the Laplacian domain, which captures both sharp and smooth image features, across scales, more robustly than previous image- and gradient-domain fitting strategies. The sparse set of curves generated by our method accurately reconstructs images and often closely matches tediously hand-authored curve data. Also, our hierarchical curves are readily usable in all existing editing frameworks. We validate our method on a broad class of images, including natural images, synthesized images with turbulent multi-scale details, and traditional vector-art, as well as illustrating simple multi-scale abstraction and color editing results.
@article{DBLP:journals/tog/Xie0TN14,
author = {Guofu Xie and
Xin Sun and
Xin Tong and
Derek Nowrouzezahrai},
title = {Hierarchical diffusion curves for accurate automatic image vectorization},
journal = {{ACM} Trans. Graph.},
volume = {33},
number = {6},
pages = {230:1--230:11},
year = {2014},
url = {https://doi.org/10.1145/2661229.2661275},
doi = {10.1145/2661229.2661275},
timestamp = {Mon, 26 Jun 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/Xie0TN14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Filtering Non-Linear TransferFunctions on Surfaces.
Heitz, E.; Nowrouzezahrai, D.; Poulin, P.; and Neyret, F.
IEEE Trans. Vis. Comput. Graph., 20(7): 996–1008. 2014.
Paper (PDF) Supplement (PDF)
Paper
doi
link
bibtex
abstract Applying non-linear transfer functions and look-up tables to procedural functions (such as noise), surface attributes, or even surface geometry are common strategies used to enhance visual detail. Their simplicity and ability to mimic a wide range of realistic appearances have led to their adoption in many rendering problems. As with any textured or geometric detail, proper filtering is needed to reduce aliasing when viewed across a range of distances, but accurate and efficient transfer function filtering remains an open problem for several reasons: transfer functions are complex and non-linear, especially when mapped through procedural noise and/or geometry-dependent functions, and the effects of perspective and masking further complicate the filtering over a pixel's footprint. We accurately solve this problem by computing and sampling from specialized filtering distributions on the fly, yielding very fast performance. We investigate the case where the transfer function to filter is a color map applied to (macroscale) surface textures (like noise), as well as color maps applied according to (microscale) geometric details. We introduce a novel representation of a (potentially modulated) color map's distribution over pixel footprints using Gaussian statistics and, in the more complex case of high-resolution color mapped microsurface details, our filtering is view- and light-dependent, and capable of correctly handling masking and occlusion effects. Our approach can be generalized to filter other physical-based rendering quantities. We propose an application to shading with irradiance environment maps over large terrains. Our framework is also compatible with the case of transfer functions used to warp surface geometry, as long as the transformations can be represented with Gaussian statistics, leading to proper view- and light-dependent filtering results. Our results match ground truth and our solution is well suited to real-time applications, requires only a few lines of shader code (provided in supplemental material), is high performance, and has a negligible memory footprint.
@article{DBLP:journals/tvcg/HeitzNPN14,
author = {Eric Heitz and
Derek Nowrouzezahrai and
Pierre Poulin and
Fabrice Neyret},
title = {Filtering Non-Linear TransferFunctions on Surfaces},
journal = {{IEEE} Trans. Vis. Comput. Graph.},
volume = {20},
number = {7},
pages = {996--1008},
year = {2014},
url = {https://doi.org/10.1109/TVCG.2013.102},
doi = {10.1109/TVCG.2013.102},
timestamp = {Wed, 14 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tvcg/HeitzNPN14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A Fast and Stable Feature-Aware Motion Blur Filter.
Guertin, J.; McGuire, M.; and Nowrouzezahrai, D.
In Wald, I.; and Ragan-Kelley, J., editor(s), High-Performance Graphics 2014, Lyon, France, 2014. Proceedings, pages 51–60, 2014. Eurographics Association
Our work was implemented as a Unity plug-in!
Paper (PDF)
Paper
doi
link
bibtex
abstract High-quality motion blur is an increasingly important effect in interactive graphics however, even in the context of offline rendering, it is often approximated as a post process. Recent motion blur post-processes (e.g., [MHBO12, Sou13]) generate plausible results with interactive performance, however distracting artifacts still remain in the presence of e.g. overlapping motion or large- and fine-scale motion features. We address these artifacts with a more robust sampling and filtering scheme with only a small additional runtime cost. We render plausible, temporally-coherent motion blur on several complex animation sequences, all in under 2ms at a resolution 1280 x 720. Moreover, our filter is designed to integrate seamlessly with post-process anti-aliasing and depth of field.
@inproceedings{DBLP:conf/egh/GuertinMN14,
author = {Jean{-}Philippe Guertin and
Morgan McGuire and
Derek Nowrouzezahrai},
editor = {Ingo Wald and
Jonathan Ragan{-}Kelley},
title = {A Fast and Stable Feature-Aware Motion Blur Filter},
booktitle = {High-Performance Graphics 2014, Lyon, France, 2014. Proceedings},
pages = {51--60},
publisher = {Eurographics Association},
year = {2014},
url = {https://doi.org/10.2312/hpg.20141093},
doi = {10.2312/HPG.20141093},
timestamp = {Wed, 24 May 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/egh/GuertinMN14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
State of the Art in Artistic Editing of Appearance, Lighting, and Material.
Schmidt, T.; Pellacini, F.; Nowrouzezahrai, D.; Jarosz, W.; and Dachsbacher, C.
In Lefebvre, S.; and Spagnuolo, M., editor(s), 35th Annual Conference of the European Association for Computer Graphics, Eurographics 2014 - State of the Art Reports, Strasbourg, France, April 7-11, 2014, pages 187–198, 2014. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract Mimicking the appearance of the real world is a longstanding goal of computer graphics, with several important applications in the feature-film, architecture and medical industries. Images with well-designed shading are an important tool for conveying information about the world, be it the shape and function of a CAD model, or the mood of a movie sequence. However, authoring this content is often a tedious task, even if undertaken by groups of highly-trained and experienced artists. Unsurprisingly, numerous methods to facilitate and accelerate this appearance editing task have been proposed, enabling the editing of scene objects' appearances, lighting, and materials, as well as entailing the introduction of new interaction paradigms and specialized preview rendering techniques. In this review we provide a comprehensive survey of artistic appearance, lighting, and material editing approaches. We organize this complex and active research area in a structure tailored to academic researchers, graduate students, and industry professionals alike. In addition to editing approaches, we discuss how user interaction paradigms and rendering backends combine to form usable systems for appearance editing. We conclude with a discussion of open problems and challenges to motivate and guide future research.
@inproceedings{DBLP:conf/eurographics/SchmidtPNJD14,
author = {Thorsten{-}Walther Schmidt and
Fabio Pellacini and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Carsten Dachsbacher},
editor = {Sylvain Lefebvre and
Michela Spagnuolo},
title = {State of the Art in Artistic Editing of Appearance, Lighting, and
Material},
booktitle = {35th Annual Conference of the European Association for Computer Graphics,
Eurographics 2014 - State of the Art Reports, Strasbourg, France,
April 7-11, 2014},
pages = {187--198},
publisher = {Eurographics Association},
year = {2014},
url = {https://doi.org/10.2312/egst.20141041},
doi = {10.2312/EGST.20141041},
timestamp = {Thu, 02 Jul 2020 13:59:55 +0200},
biburl = {https://dblp.org/rec/conf/eurographics/SchmidtPNJD14.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2013
(6)
|
Image-based reconstruction and synthesis of dense foliage.
Bradley, D.; Nowrouzezahrai, D.; and Beardsley, P. A.
ACM Trans. Graph., 32(4): 74:1–74:10. 2013.
Paper (PDF)
Paper
doi
link
bibtex
2 downloads
abstract Flora is an element in many computer-generated scenes. But trees, bushes and plants have complex geometry and appearance, and are difficult to model manually. One way to address this is to capture models directly from the real world. Existing techniques have focused on extracting macro structure such as the branching structure of trees, or the structure of broad-leaved plants with a relatively small number of surfaces. This paper presents a finer scale technique to demonstrate for the first time the processing of densely leaved foliage - computation of 3D structure, plus extraction of statistics for leaf shape and the configuration of neighboring leaves. Our method starts with a mesh of a single exemplar leaf of the target foliage. Using a small number of images, point cloud data is obtained from multi-view stereo, and the exemplar leaf mesh is fitted non-rigidly to the point cloud over several iterations. Initialization of the fitting is done using RANSAC, making it robust to outliers in the stereo reconstruction and suitable for the chaotic point cloud obtained from foliage. In addition, our method learns a statistical model of leaf shape and appearance during the reconstruction phase, and a model of the transformations between neighboring leaves. This information can subsequently be used to generate a variety of plausible leaves for that plant species in plausible configuration, and is useful in two ways - to augment and increase leaf density in reconstructions of captured foliage, and to synthesize new foliage that conforms to a user-specified layout and density. The result of our technique is a dense set of captured leaves with realistic appearance, and a method for leaf synthesis. Our approach excels at reconstructing plants and bushes that are primarily defined by dense leaves and is demonstrated with multiple examples.
@article{DBLP:journals/tog/BradleyNB13,
author = {Derek Bradley and
Derek Nowrouzezahrai and
Paul A. Beardsley},
title = {Image-based reconstruction and synthesis of dense foliage},
journal = {{ACM} Trans. Graph.},
volume = {32},
number = {4},
pages = {74:1--74:10},
year = {2013},
url = {https://doi.org/10.1145/2461912.2461952},
doi = {10.1145/2461912.2461952},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/BradleyNB13.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Path-space manipulation of physically-based light transport.
Schmidt, T.; Novák, J.; Meng, J.; Kaplanyan, A.; Reiner, T.; Nowrouzezahrai, D.; and Dachsbacher, C.
ACM Trans. Graph., 32(4): 129:1–129:11. 2013.
Lightrig spin-off company
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract Industry-quality content creation relies on tools for lighting artists to quickly prototype, iterate, and refine final renders. As industryleading studios quickly adopt physically-based rendering (PBR) across their art generation pipelines, many existing tools have become unsuitable as they address only simple effects without considering underlying PBR concepts and constraints. We present a novel light transport manipulation technique that operates directly on path-space solutions of the rendering equation. We expose intuitive direct and indirect manipulation approaches to edit complex effectssuch as (multi-refracted) caustics, diffuse and glossy indirect bounces, and direct / indirect shadows. With our sketch- and objectspace selection, all built atop a parameterized regular expression engine, artists can search and isolate shading effects to inspect and edit. We classify and filter paths on the fly and visualize the selected transport phenomena. We survey artists who used our tool to manipulate complex phenomena on both static and animated scenes.
@article{DBLP:journals/tog/SchmidtNMKRND13,
author = {Thorsten{-}Walther Schmidt and
Jan Nov{\'{a}}k and
Johannes Meng and
Anton Kaplanyan and
Tim Reiner and
Derek Nowrouzezahrai and
Carsten Dachsbacher},
title = {Path-space manipulation of physically-based light transport},
journal = {{ACM} Trans. Graph.},
volume = {32},
number = {4},
pages = {129:1--129:11},
year = {2013},
url = {https://doi.org/10.1145/2461912.2461980},
doi = {10.1145/2461912.2461980},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/SchmidtNMKRND13.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Joint importance sampling of low-order volumetric scattering.
Georgiev, I.; Krivánek, J.; Hachisuka, T.; Nowrouzezahrai, D.; and Jarosz, W.
ACM Trans. Graph., 32(6): 164:1–164:14. 2013.
Paper (PDF)
Paper
doi
link
bibtex
abstract Central to all Monte Carlo-based rendering algorithms is the construction of light transport paths from the light sources to the eye. Existing rendering approaches sample path vertices incrementally when constructing these light transport paths. Paths should ideally be constructed according to a joint probability density function proportional to the integrand, yet current incremental sampling strategies only locally account for certain terms in the integrand. The resulting probability density is thus a product of the conditional densities of each local sampling step, constructed without explicit control over the form of the final joint distribution of the complete path. We analyze why current incremental construction schemes often lead to high variance in the presence of participating media, and reveal that such approaches are an unnecessary legacy inherited from traditional surface-based rendering algorithms. We devise joint importance sampling of path vertices in participating media to construct paths that explicitly account for the product of all scattering and geometry terms along a sequence of vertices instead of just locally at a single vertex. This leads to a number of practical importance sampling routines to explicitly construct single- and double-scattering subpaths in anisotropically-scattering media. We demonstrate the benefit of our new sampling techniques, integrating them into several path-based rendering algorithms such as path tracing, bidirectional path tracing, and many-light methods. We also use our sampling routines to generalize deterministic shadow connections to connection subpaths consisting of two or three random decisions, to efficiently simulate higher-order multiple scattering. Our algorithms significantly reduce noise and increase performance in renderings with both isotropic and highly anisotropic, low-order scattering.
@article{DBLP:journals/tog/GeorgievKHNJ13,
author = {Iliyan Georgiev and
Jaroslav Kriv{\'{a}}nek and
Toshiya Hachisuka and
Derek Nowrouzezahrai and
Wojciech Jarosz},
title = {Joint importance sampling of low-order volumetric scattering},
journal = {{ACM} Trans. Graph.},
volume = {32},
number = {6},
pages = {164:1--164:14},
year = {2013},
url = {https://doi.org/10.1145/2508363.2508411},
doi = {10.1145/2508363.2508411},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/GeorgievKHNJ13.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Filtering color mapped textures and surfaces.
Heitz, E.; Nowrouzezahrai, D.; Poulin, P.; and Neyret, F.
In Gopi, M.; Yoon, S.; Spencer, S. N.; Olano, M.; and Otaduy, M. A., editor(s), Symposium on Interactive 3D Graphics and Games, I3D '13, Orlando, FL, USA, March 22-24, 2013, pages 129–136, 2013. ACM
Best Paper Award Winner!
Paper (PDF) Supplement (PDF)
Paper
doi
link
bibtex
abstract Color map textures applied directly to surfaces, to geometric microsurface details, or to procedural functions (such as noise), are commonly used to enhance visual detail. Their simplicity and ability to mimic a wide range of realistic appearances have led to their adoption in many rendering problems. As with any textured or geometric detail, proper filtering is needed to reduce aliasing when viewed across a range of distances, but accurate and efficient color map filtering remains an open problem for several reasons: color maps are complex non-linear functions, especially when mapped through procedural noise and/or geometry-dependent functions, and the effects of perspective and masking further complicate the filtering over a pixel's footprint. We accurately solve this problem by computing and sampling from specialized filtering distributions on-thefly, yielding very fast performance. We filter color map textures applied to (macro-scale) surfaces, as well as color maps applied according to (micro-scale) geometric details. We introduce a novel representation of a (potentially modulated) color map's distribution over pixel footprints using Gaussian statistics and, in the more complex case of high-resolution color mapped microsurface details, our filtering is view- and light-dependent, and capable of correctly handling masking and occlusion effects. Our results match ground truth and our solution is well suited to real-time applications, requires only a few lines of shader code (provided in supplemental material), is high performance, and has a negligible memory footprint.
@inproceedings{DBLP:conf/si3d/HeitzNPN13,
author = {Eric Heitz and
Derek Nowrouzezahrai and
Pierre Poulin and
Fabrice Neyret},
editor = {Meenakshisundaram Gopi and
Sung{-}Eui Yoon and
Stephen N. Spencer and
Marc Olano and
Miguel A. Otaduy},
title = {Filtering color mapped textures and surfaces},
booktitle = {Symposium on Interactive 3D Graphics and Games, {I3D} '13, Orlando,
FL, USA, March 22-24, 2013},
pages = {129--136},
publisher = {{ACM}},
year = {2013},
url = {https://doi.org/10.1145/2448196.2448217},
doi = {10.1145/2448196.2448217},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/si3d/HeitzNPN13.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
State of the art in photon density estimation.
Hachisuka, T.; Jarosz, W.; Georgiev, I.; Kaplanyan, A.; Nowrouzezahrai, D.; and Spencer, B.
In Yu, Y., editor(s), SIGGRAPH Asia 2013, Hong Kong, China, November 19-22, 2013, Courses, pages 15:1–15:562, 2013. ACM
Paper
doi
link
bibtex
abstract Photon density estimation techniques are a popular choice for simulating light transport in scenes with complicated geometry and materials. This class of algorithms can be used to accurately simulate inter-reflections, caustics, color bleeding, scattering in participating media, and subsurface scattering. Since its introduction, photon density estimation has been significantly extended in computer graphics with the introduction of: specialized techniques that intelligently modify the positions or bandwidths to reduce visual error using a small number of photons, approaches which eliminate error completely in the limit, methods that use higher-order samples and queries to reduce error in participating media, and recent generalized formulations that bridge the gap between photon density estimation and other techniques. This course provides the necessary insight to implement all these latest advances in photon density estimation. The course starts out with a short introduction to photon density estimation using classical photon mapping, but the remainder of the two-part course provides hands-on explanations of the latest developments in this area by the experts behind each technique. The course will give the audience concrete and practical understanding of the latest developments in photon density estimation techniques that have not been presented in prior similar SIGGRAPH/SIGGRAPH Asia courses.
@inproceedings{DBLP:conf/siggraph/HachisukaJGKNS13,
author = {Toshiya Hachisuka and
Wojciech Jarosz and
Iliyan Georgiev and
Anton Kaplanyan and
Derek Nowrouzezahrai and
Ben Spencer},
editor = {Yizhou Yu},
title = {State of the art in photon density estimation},
booktitle = {{SIGGRAPH} Asia 2013, Hong Kong, China, November 19-22, 2013, Courses},
pages = {15:1--15:562},
publisher = {{ACM}},
year = {2013},
url = {https://doi.org/10.1145/2542266.2542281},
doi = {10.1145/2542266.2542281},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/siggraph/HachisukaJGKNS13.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2012
(9)
|
Progressive Virtual Beam Lights.
Novák, J.; Nowrouzezahrai, D.; Dachsbacher, C.; and Jarosz, W.
Comput. Graph. Forum, 31(4): 1407–1413. 2012.
Featured on Proceedings Back Cover!
Paper (PDF)
Paper
doi
link
bibtex
abstract A recent technique that forms virtual ray lights (VRLs) from path segments in media, reduces the artifacts common to VPL approaches in participating media, however, distracting singularities still remain. We present Virtual Beam Lights (VBLs), a progressive many-lights algorithm for rendering complex indirect transport paths in, from, and to media. VBLs are efficient and can handle heterogeneous media, anisotropic scattering, and moderately glossy surfaces, while provably converging to ground truth. We inflate ray lights into beam lights with finite thicknesses to eliminate the remaining singularities. Furthermore, we devise several practical schemes for importance sampling the various transport contributions between camera rays, light rays, and surface points. VBLs produce artifact-free images faster than VRLs, especially when glossy surfaces and/or anisotropic phase functions are present. Lastly, we employ a progressive thickness reduction scheme for VBLs in order to render results that converge to ground truth.
@article{DBLP:journals/cgf/NovakNDJ12,
author = {Jan Nov{\'{a}}k and
Derek Nowrouzezahrai and
Carsten Dachsbacher and
Wojciech Jarosz},
title = {Progressive Virtual Beam Lights},
journal = {Comput. Graph. Forum},
volume = {31},
number = {4},
pages = {1407--1413},
year = {2012},
url = {https://doi.org/10.1111/j.1467-8659.2012.03136.x},
doi = {10.1111/J.1467-8659.2012.03136.X},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/NovakNDJ12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Irradiance Rigs.
Yuan, H.; Nowrouzezahrai, D.; and Sloan, P. J.
J. Graph. Tools, 16(1): 1–11. 2012.
Paper
doi
link
bibtex
5 downloads
abstract When precomputed lighting is generated for static scene elements, the incident illumination on dynamic objects must be computed in a manner that is efficient and that faithfully captures the near- and far-field variation of the environment's illumination. Depending on the relative size of dynamic objects, as well as the number of lights in the scene, previous approaches fail to adequately sample the incident lighting and/or fail to scale. We present a principled, error-driven approach for dynamically transitioning between near- and far-field lighting. A more accurate model for sampling near-field lighting for disk sources is introduced, as well as far-field sampling and interpolation schemes tailored to each dynamic object. Lastly, we apply a flexible reflectance model to the computed illumination.
@article{DBLP:journals/jgtools/YuanNS12,
author = {Hong Yuan and
Derek Nowrouzezahrai and
Peter{-}Pike J. Sloan},
title = {Irradiance Rigs},
journal = {J. Graph. Tools},
volume = {16},
number = {1},
pages = {1--11},
year = {2012},
url = {https://doi.org/10.1080/2151237X.2012.631448},
doi = {10.1080/2151237X.2012.631448},
timestamp = {Thu, 18 Jun 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/jgtools/YuanNS12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Learning hatching for pen-and-ink illustration of surfaces.
Kalogerakis, E.; Nowrouzezahrai, D.; Breslav, S.; and Hertzmann, A.
ACM Trans. Graph., 31(1): 1:1–1:17. 2012.
Paper (PDF)
Paper
doi
link
bibtex
abstract This paper presents an algorithm for learning hatching styles from line drawings. An artist draws a single hatching illustration of a 3D object. Their strokes are analyzed to extract the following per-pixel properties: hatching level (hatching, cross-hatching, or no strokes), stroke orientation, spacing, intensity, length, and thickness. A mapping is learned from input features to these properties, using classification, regression, and clustering techniques. Then, a new illustration can be generated in the artist's style, as follows. First, given a new view of a 3D object, the learned mapping is applied to synthesize target stroke properties for each pixel. A new illustration is then generated by synthesizing hatching strokes according to the target properties.
@article{DBLP:journals/tog/KalogerakisNBH12,
author = {Evangelos Kalogerakis and
Derek Nowrouzezahrai and
Simon Breslav and
Aaron Hertzmann},
title = {Learning hatching for pen-and-ink illustration of surfaces},
journal = {{ACM} Trans. Graph.},
volume = {31},
number = {1},
pages = {1:1--1:17},
year = {2012},
url = {https://doi.org/10.1145/2077341.2077342},
doi = {10.1145/2077341.2077342},
timestamp = {Sun, 22 Oct 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/KalogerakisNBH12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Sparse zonal harmonic factorization for efficient SH rotation.
Nowrouzezahrai, D.; Simari, P. D.; and Fiume, E.
ACM Trans. Graph., 31(3): 23:1–23:9. 2012.
Paper (PDF) Supplement (PDF)
Paper
doi
link
bibtex
abstract We present a sparse analytic representation for spherical functions, including those expressed in a spherical harmonic (SH) expansion, that is amenable to fast and accurate rotation on the GPU. Exploiting the fact that each band-l SH basis function can be expressed as a weighted sum of 2l+1 rotated band-l zonal harmonic (ZH) lobes, we develop a factorization that significantly reduces this number. We investigate approaches for promoting sparsity in the change-of-basis matrix, and also introduce lobe sharing to reduce the total number of unique lobe directions used for an order-N expansion from N*N to 2N-1. Our representation does not introduce approximation error, is suitable for any type of spherical function (e.g., lighting or transfer), and requires no offline fitting procedure; only a (sparse) matrix multiplication is required to map to/from SH. We provide code for our rotation algorithms, and apply them to several real-time rendering applications.
@article{DBLP:journals/tog/NowrouzezahraiSF12,
author = {Derek Nowrouzezahrai and
Patricio D. Simari and
Eugene Fiume},
title = {Sparse zonal harmonic factorization for efficient {SH} rotation},
journal = {{ACM} Trans. Graph.},
volume = {31},
number = {3},
pages = {23:1--23:9},
year = {2012},
url = {https://doi.org/10.1145/2167076.2167081},
doi = {10.1145/2167076.2167081},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/NowrouzezahraiSF12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A theory of monte carlo visibility sampling.
Ramamoorthi, R.; Anderson, J.; Meyer, M.; and Nowrouzezahrai, D.
ACM Trans. Graph., 31(5): 121:1–121:16. 2012.
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract Soft shadows from area lights are one of the most crucial effects in high quality and production rendering, but Monte Carlo sampling of visibility is often the main source of noise in rendered images. Indeed, it is common to use deterministic uniform sampling for the smoother shading effects in direct lighting, so that all of the Monte-Carlo noise arises from visibility sampling alone. In this paper, we analyze theoretically and empirically, using both statistical and Fourier methods, the effectiveness of different non-adaptive Monte Carlo sampling patterns for rendering soft shadows. We start with a single image scanline and a linear light source, and gradually consider more complex visibility functions at a pixel. We show analytically that the lowest expected variance is in fact achieved by uniform sampling (albeit at the cost of visual banding artifacts). Surprisingly, we show that for two or more discontinuities in the visibility function, a comparable error to uniform sampling is obtained by uniform jitter sampling, where a constant jitter is applied to all samples in a uniform pattern (as opposed to jittering each stratum as in standard stratified sampling). The variance can be reduced by up to a factor of two, compared to stratified or quasi-Monte Carlo techniques, without the banding in uniform sampling. We augment our statistical analysis with a novel 2D Fourier analysis across the pixel-light space. This allows us to characterize the banding frequencies in uniform sampling, and gives insights into the behavior of uniform jitter and stratified sampling. We next extend these results to planar area light sources. We show that the best sampling method can vary, depending on the type of light source (circular, gaussian or square/rectangular). The correlation of adjacent light scanlines in square light sources can reduce the effectiveness of uniform jitter sampling, while the smoother shape of circular and gaussian-modulated sources preserves its benefits - these findings are also exposed through our frequency analysis. In practical terms, the theory in this paper provides guidelines for selecting visibility sampling strategies, which can reduce the number of shadow samples by 20 to 40 percent, with simple modifications to existing rendering code.
@article{DBLP:journals/tog/RamamoorthiAMN12,
author = {Ravi Ramamoorthi and
John Anderson and
Mark Meyer and
Derek Nowrouzezahrai},
title = {A theory of monte carlo visibility sampling},
journal = {{ACM} Trans. Graph.},
volume = {31},
number = {5},
pages = {121:1--121:16},
year = {2012},
url = {https://doi.org/10.1145/2231816.2231819},
doi = {10.1145/2231816.2231819},
timestamp = {Tue, 25 Aug 2020 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/RamamoorthiAMN12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
The magic lens: refractive steganography.
Papas, M.; Houit, T.; Nowrouzezahrai, D.; Gross, M. H.; and Jarosz, W.
ACM Trans. Graph., 31(6): 186:1–186:10. 2012.
Featured on Proceedings Back Cover and Papers Fast Forward Video!
Paper (PDF) Supplement (PDF)
Paper
doi
link
bibtex
2 downloads
abstract We present an automatic approach to design and manufacture passive display devices based on optical hidden image decoding. Motivated by classical steganography techniques we construct Magic Lenses, composed of refractive lenslet arrays, to reveal hidden images when placed over potentially unstructured printed or displayed source images. We determine the refractive geometry of these surfaces by formulating and efficiently solving an inverse light transport problem, taking into account additional constraints imposed by physical manufacturing processes. We fabricate several variants on the basic magic lens idea including using a single source image to encode several hidden images which are only revealed when the lens is placed at prescribed rotational orientations or viewed from different angles. We also present an important special case, the universal lens, that forms an injunction with the source image grid and can be applied to arbitrary source images. We use this type of lens to generate hidden animation sequences. We validate our simulation results with many real-world manufactured magic lenses, and experiment with two separate manufacturing processes.
@article{DBLP:journals/tog/PapasHNGJ12,
author = {Marios Papas and
Thomas Houit and
Derek Nowrouzezahrai and
Markus H. Gross and
Wojciech Jarosz},
title = {The magic lens: refractive steganography},
journal = {{ACM} Trans. Graph.},
volume = {31},
number = {6},
pages = {186:1--186:10},
year = {2012},
url = {https://doi.org/10.1145/2366145.2366205},
doi = {10.1145/2366145.2366205},
timestamp = {Tue, 07 May 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/PapasHNGJ12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Delta radiance transfer.
Loos, B. J.; Nowrouzezahrai, D.; Jarosz, W.; and Sloan, P. J.
In Garland, M.; Wang, R.; Spencer, S. N.; Gopi, M.; and Yoon, S., editor(s), Symposium on Interactive 3D Graphics and Games, I3D '12, Costa Mesa, CA, USA, March 09 - 11, 2012, pages 191–196, 2012. ACM
Paper (PDF)
Paper
doi
link
bibtex
2 downloads
abstract Modular Radiance Transfer (MRT) is a recent technique for computing approximate direct-to-indirect transport. Scenes are dynamically constructed by warping and connecting simple shapes and compact transport operators are only precomputed on these simple shapes. MRT ignores fine-scale transport from clutter objects inside the scene, and computes light transport with reduced dimensional operators, which allows extremely high performance but can lead to significant approximation error. We present several techniques to alleviate this limitation, allowing the light transport from clutter in a scene to be accounted for. We derive additional low-rank delta operators to compensate for these missing light transport paths by modeling indirect shadows and interreflections from, and onto, clutter objects in the scene. We retain MRT's scene-independent precomputation and augment its scene-dependent initialization with clutter transport generation, resulting in increased accuracy without a performance penalty. Our implementation is simple, requiring a few small matrix-vector multiplications that generate a delta lightmap added to MRT's output, and does not adversely affect the performance benefits of the overall algorithm.
@inproceedings{DBLP:conf/si3d/LoosNJS12,
author = {Bradford James Loos and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Peter{-}Pike J. Sloan},
editor = {Michael Garland and
Rui Wang and
Stephen N. Spencer and
Meenakshisundaram Gopi and
Sung{-}Eui Yoon},
title = {Delta radiance transfer},
booktitle = {Symposium on Interactive 3D Graphics and Games, {I3D} '12, Costa Mesa,
CA, USA, March 09 - 11, 2012},
pages = {191--196},
publisher = {{ACM}},
year = {2012},
url = {https://doi.org/10.1145/2159616.2159648},
doi = {10.1145/2159616.2159648},
timestamp = {Tue, 06 Nov 2018 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/si3d/LoosNJS12.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2011
(9)
|
Exploiting Coherence and Data-driven Models for Real-time Global Illumination.
Nowrouzezahrai, D.
Ph.D. Thesis, University of Toronto, Canada, 2011.
Paper
link
bibtex
@phdthesis{DBLP:phd/ca/Nowrouzezahrai11,
author = {Derek Nowrouzezahrai},
title = {Exploiting Coherence and Data-driven Models for Real-time Global Illumination},
school = {University of Toronto, Canada},
year = {2011},
url = {http://hdl.handle.net/1807/26218},
timestamp = {Fri, 29 Jul 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/phd/ca/Nowrouzezahrai11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Wrap Shading.
Sloan, P. J.; Nowrouzezahrai, D.; and Yuan, H.
J. Graphics, GPU, & Game Tools, 15(4): 252–259. 2011.
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract Shading models that wrap around the hemisphere have been used to approximate subsurface scattering, area light sources or just as a softer re?ectance model. We present a generalization of a technique that has been successfully used in games and include details on how to incorporate them when lighting using spherical harmonics.
@article{DBLP:journals/jgtools/SloanNY11,
author = {Peter{-}Pike J. Sloan and
Derek Nowrouzezahrai and
Hong Yuan},
title = {Wrap Shading},
journal = {J. Graphics, GPU, {\&} Game Tools},
volume = {15},
number = {4},
pages = {252--259},
year = {2011},
url = {https://doi.org/10.1080/2151237X.2011.628841},
doi = {10.1080/2151237X.2011.628841},
timestamp = {Thu, 18 May 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/jgtools/SloanNY11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A comprehensive theory of volumetric radiance estimation using photon points and beams.
Jarosz, W.; Nowrouzezahrai, D.; Sadeghi, I.; and Jensen, H. W.
ACM Trans. Graph., 30(1): 5:1–5:19. 2011.
Presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
abstract We present two contributions to the area of volumetric rendering. We develop a novel, comprehensive theory of volumetric radiance estimation that leads to several new insights and includes all previously published estimates as special cases. This theory allows for estimating in-scattered radiance at a point, or accumulated radiance along a camera ray, with the standard photon particle representation used in previous work. Furthermore, we generalize these operations to include a more compact, and more expressive intermediate representation of lighting in participating media, which we call ``photon beams.'' The combination of these representations and their respective query operations results in a collection of nine distinct volumetric radiance estimates. Our second contribution is a more efficient rendering method for participating media based on photon beams. Even when shooting and storing less photons and using less computation time, our method significantly reduces both bias (blur) and variance in volumetric radiance estimation. This enables us to render sharp lighting details (e.g. volume caustics) using just tens of thousands of photon beams, instead of the millions to billions of photon points required with previous methods.
@article{DBLP:journals/tog/JaroszNSJ11,
author = {Wojciech Jarosz and
Derek Nowrouzezahrai and
Iman Sadeghi and
Henrik Wann Jensen},
title = {A comprehensive theory of volumetric radiance estimation using photon
points and beams},
journal = {{ACM} Trans. Graph.},
volume = {30},
number = {1},
pages = {5:1--5:19},
year = {2011},
url = {https://doi.org/10.1145/1899404.1899409},
doi = {10.1145/1899404.1899409},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/JaroszNSJ11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
A programmable system for artistic volumetric lighting.
Nowrouzezahrai, D.; Johnson, J. M.; Selle, A.; Lacewell, D.; Kaschalk, M.; and Jarosz, W.
ACM Trans. Graph., 30(4): 29. 2011.
Featured on Proceedings Back Cover and in Papers Fast Forward Video!
Paper (PDF)
Paper
doi
link
bibtex
4 downloads
abstract We present a method for generating art-directable volumetric effects, ranging from physically-accurate to non-physical results. Our system mimics the way experienced artists think about volumetric effects by using an intuitive lighting primitive, and decoupling the modeling and shading of this primitive. To accomplish this, we generalize the physically-based photon beams method to allow arbitrarily programmable simulation and shading phases. This provides an intuitive design space for artists to rapidly explore a wide range of physically-based as well as plausible, but exaggerated, volumetric effects. We integrate our approach into a real-world production pipeline and couple our volumetric effects to surface shading.
@article{DBLP:journals/tog/NowrouzezahraiJSLKJ11,
author = {Derek Nowrouzezahrai and
Jared M. Johnson and
Andrew Selle and
Dylan Lacewell and
Michael Kaschalk and
Wojciech Jarosz},
title = {A programmable system for artistic volumetric lighting},
journal = {{ACM} Trans. Graph.},
volume = {30},
number = {4},
pages = {29},
year = {2011},
url = {https://doi.org/10.1145/2010324.1964924},
doi = {10.1145/2010324.1964924},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/NowrouzezahraiJSLKJ11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Modular Radiance Transfer.
Loos, B. J.; Antani, L.; Mitchell, K.; Nowrouzezahrai, D.; Jarosz, W.; and Sloan, P. J.
ACM Trans. Graph., 30(6): 178. 2011.
Featured in Papers Fast Forward Video!
Paper (PDF)
Paper
doi
link
bibtex
abstract Many rendering algorithms willingly sacrifice accuracy, favoring plausible shading with high-performance. We present Modular Radiance Transfer (MRT), an approach for smooth, approximate direct-to-indirect transfer that scales from high-end GPUs to low-end mobile platforms. MRT eliminates scene-dependent precomputation by storing compact transport on simple shapes, akin to bounce cards used in film production. These shapes' modular transport can be instanced, warped and connected on-the-fly to yield approximate light transport in large scenes. We introduce a prior on incident lighting distributions and perform all computations in low-dimensional subspaces. An {implicit lighting environment} induced from the low-rank approximations is in turn used to model secondary effects, such as volumetric transport variation, higher-order irradiance, and transport through lightfields. MRT is a new approach to precomputed lighting that uses a novel low-dimensional subspace simulation of light transport to uniquely balance the need for high-performance and portable solutions, low memory usage, and fast authoring iteration.
@article{DBLP:journals/tog/LoosAMNJS11,
author = {Bradford James Loos and
Lakulish Antani and
Kenny Mitchell and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Peter{-}Pike J. Sloan},
title = {Modular Radiance Transfer},
journal = {{ACM} Trans. Graph.},
volume = {30},
number = {6},
pages = {178},
year = {2011},
url = {https://doi.org/10.1145/2070781.2024212},
doi = {10.1145/2070781.2024212},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/LoosAMNJS11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Progressive photon beams.
Jarosz, W.; Nowrouzezahrai, D.; Thomas, R.; Sloan, P. J.; and Zwicker, M.
ACM Trans. Graph., 30(6): 181. 2011.
Featured in the Proceedings Inside Cover and the Papers Fast Forward Video!
Paper (PDF)
Paper
doi
link
bibtex
abstract We present progressive photon beams, a new algorithm for rendering complex lighting in participating media. Our technique is efficient, robust to complex light paths, and handles heterogeneous media and anisotropic scattering while provably converging to the correct solution using a bounded memory footprint. We achieve this by extending the recent photon beams variant of volumetric photon mapping. We show how to formulate a progressive radiance estimate using photon beams, providing the convergence guarantees and bounded memory usage of progressive photon mapping. Progressive photon beams can robustly handle situations that are difficult for most other algorithms, such as scenes containing participating media and specular interfaces, with realistic light sources completely enclosed by refractive and reflective materials. Our technique handles heterogeneous media and also trivially supports stochastic effects such as depth-of-field and glossy materials. Finally, we show how progressive photon beams can be implemented efficiently on the GPU as a splatting operation, making it applicable to interactive and real-time applications. These features make our technique scalable, providing the same physically-based algorithm for interactive feedback and reference-quality, unbiased solutions.
@article{DBLP:journals/tog/JaroszNTSZ11,
author = {Wojciech Jarosz and
Derek Nowrouzezahrai and
Robert Thomas and
Peter{-}Pike J. Sloan and
Matthias Zwicker},
title = {Progressive photon beams},
journal = {{ACM} Trans. Graph.},
volume = {30},
number = {6},
pages = {181},
year = {2011},
url = {https://doi.org/10.1145/2070781.2024215},
doi = {10.1145/2070781.2024215},
timestamp = {Sun, 02 Oct 2022 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/tog/JaroszNTSZ11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Perceptually-based compensation of light pollution in display systems.
van Baar, J.; Poulakos, S.; Jarosz, W.; Nowrouzezahrai, D.; Tamstorf, R.; and Gross, M. H.
In McDonnell, R.; Thorpe, S. J.; Spencer, S. N.; Gutierrez, D.; and Giese, M. A., editor(s), Proceedings of the 8th Symposium on Applied Perception in Graphics and Visualization, APGV 2011, Toulouse, France, August 27-28, 2011, pages 45–52, 2011. ACM
Paper (PDF)
Paper
doi
link
bibtex
abstract This paper addresses the problem of unintended light contributions due to physical properties of display systems. An example of such unintended contribution is crosstalk in stereoscopic 3D display systems, often referred to as ghosting. Ghosting results in a reduction of visual quality, and may lead to an uncomfortable viewing experience. The latter is due to conflicting (depth) edge cues, which can hinder the human visual system (HVS) proper fusion of stereo images (stereopsis). We propose an automatic, perceptually-based computational compensation framework, which formulates pollution elimination as a minimization problem. Our method aims to distribute the error introduced by the pollution in a perceptually optimal manner. As a consequence ghost edges are smoothed locally, resulting in a more comfortable stereo viewing experience. We show how to make the computation tractable by exploiting the structure of the resulting problem, and also propose a perceptually-based pollution prediction. We show that our general framework is applicable to other light pollution problems, such as descattering.
@inproceedings{DBLP:conf/apgv/BaarPJNTG11,
author = {Jeroen van Baar and
Steven Poulakos and
Wojciech Jarosz and
Derek Nowrouzezahrai and
Rasmus Tamstorf and
Markus H. Gross},
editor = {Rachel McDonnell and
Simon J. Thorpe and
Stephen N. Spencer and
Diego Gutierrez and
Martin A. Giese},
title = {Perceptually-based compensation of light pollution in display systems},
booktitle = {Proceedings of the 8th Symposium on Applied Perception in Graphics
and Visualization, {APGV} 2011, Toulouse, France, August 27-28, 2011},
pages = {45--52},
publisher = {{ACM}},
year = {2011},
url = {https://doi.org/10.1145/2077451.2077460},
doi = {10.1145/2077451.2077460},
timestamp = {Tue, 07 May 2024 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/apgv/BaarPJNTG11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Runtime implementation of modular radiance transfer.
Loos, B.; Antani, L.; Mitchell, K.; Nowrouzezahrai, D.; Jarosz, W.; and Sloan, P. J.
In Elendt, M., editor(s), International Conference on Computer Graphics and Interactive Techniques, SIGGRAPH 2011, Vancouver, BC, Canada, August 7-11, 2011, Talks Proceedings, pages 59, 2011. ACM
Paper
doi
link
bibtex
1 download
abstract Real-time rendering of indirect lighting significantly enhances the sense of realism in video games. Unfortunately, previously including such effects often required time consuming scene dependent precomputation and heavy runtime computations unsuitable for low-end devices, such as mobile phones or game consoles. Modular Radiance Transfer (MRT) is a recent technique that computes approximate direct-to-indirect transfer by warping and combining light transport, in real-time, from a small library of simple shapes. This talk focusses on implementation issues of the MRT technical paper, including how our run time is designed to scale across many different platforms, from iPhones to modern GPUs.
@inproceedings{DBLP:conf/siggraph/LoosAMNJS11,
author = {Brad Loos and
Lakulish Antani and
Kenny Mitchell and
Derek Nowrouzezahrai and
Wojciech Jarosz and
Peter{-}Pike J. Sloan},
editor = {Mark Elendt},
title = {Runtime implementation of modular radiance transfer},
booktitle = {International Conference on Computer Graphics and Interactive Techniques,
{SIGGRAPH} 2011, Vancouver, BC, Canada, August 7-11, 2011, Talks Proceedings},
pages = {59},
publisher = {{ACM}},
year = {2011},
url = {https://doi.org/10.1145/2037826.2037905},
doi = {10.1145/2037826.2037905},
timestamp = {Thu, 11 Mar 2021 13:39:51 +0100},
biburl = {https://dblp.org/rec/conf/siggraph/LoosAMNJS11.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2010
(1)
|
Irradiance rigs.
Yuan, H.; Nowrouzezahrai, D.; and Sloan, P. J.
In Mohler, J. L., editor(s), International Conference on Computer Graphics and Interactive Techniques, SIGGRAPH 2010, Los Angeles, California, USA, July 26-30, 2010, Talks Proceedings, 2010. ACM
Paper
doi
link
bibtex
1 download
abstract When precomputed lighting is generated for static scene elements, the incident illumination on dynamic objects must be computed in a manner that is efficient and that faithfully captures the near- and far-field variation of the environment's illumination. Depending on the relative size of dynamic objects, as well as the number of lights in the scene, previous approaches fail to adequately sample the incident lighting and/or fail to scale. We present a principled, error-driven approach for dynamically transitioning between near- and far-field lighting. A more accurate model for sampling near-field lighting for disk sources is introduced, as well as far-field sampling and interpolation schemes tailored to each dynamic object. Lastly, we apply a flexible reflectance model to the computed illumination.
@inproceedings{DBLP:conf/siggraph/YuanNS10,
author = {Hong Yuan and
Derek Nowrouzezahrai and
Peter{-}Pike J. Sloan},
editor = {James L. Mohler},
title = {Irradiance rigs},
booktitle = {International Conference on Computer Graphics and Interactive Techniques,
{SIGGRAPH} 2010, Los Angeles, California, USA, July 26-30, 2010, Talks
Proceedings},
publisher = {{ACM}},
year = {2010},
url = {https://doi.org/10.1145/1837026.1837084},
doi = {10.1145/1837026.1837084},
timestamp = {Tue, 06 Nov 2018 16:59:18 +0100},
biburl = {https://dblp.org/rec/conf/siggraph/YuanNS10.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2009
(5)
|
Shadowing Dynamic Scenes with Arbitrary BRDFs.
Nowrouzezahrai, D.; Kalogerakis, E.; and Fiume, E.
Comput. Graph. Forum, 28(2): 249–258. 2009.
Paper (PDF)
Paper
doi
link
bibtex
abstract We present a real-time relighting and shadowing method for dynamic scenes with varying lighting, view and BRDFs. Our approach is based on a compact representation of reflectance data that allows for changing the BRDF at run-time and a data-driven method for accurately synthesizing self-shadows on articulated and deformable geometries. Unlike previous self-shadowing approaches, we do not rely on local blocking heuristics. We do not fit a model to the BRDF-weighted visibility, but rather only to the visibility that changes during animation. In this manner, our model is more compact than previous techniques and requires less computation both during fitting and at run-time. Our reflectance product operators can re-integrate arbitrary low-frequency view-dependent BRDF effects on-the-fly and are compatible with all previous dynamic visibility generation techniques as well as our own data-driven visibility model. We apply our reflectance product operators to three different visibility generation models, and our data-driven model can achieve framerates well over 300Hz.
@article{DBLP:journals/cgf/NowrouzezahraiKF09,
author = {Derek Nowrouzezahrai and
Evangelos Kalogerakis and
Eugene Fiume},
title = {Shadowing Dynamic Scenes with Arbitrary BRDFs},
journal = {Comput. Graph. Forum},
volume = {28},
number = {2},
pages = {249--258},
year = {2009},
url = {https://doi.org/10.1111/j.1467-8659.2009.01364.x},
doi = {10.1111/J.1467-8659.2009.01364.X},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/cgf/NowrouzezahraiKF09.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Fast Global Illumination on Dynamic Height Fields.
Nowrouzezahrai, D.; and Snyder, J. M.
Comput. Graph. Forum, 28(4): 1131–1139. 2009.
Featured on Proceedings Back Cover!
Paper (PDF) Slides (PPT) Supplement (PDF)
Paper
doi
link
bibtex
abstract We present a real-time method for rendering global illumination effects from large area and environmental lights on dynamic height fields. In contrast to previous work, our method handles inter-reflections (indirect lighting) and non-diffuse surfaces. To reduce sampling, we construct one multi-resolution pyramid for height variation to compute direct shadows, and another pyramid for each indirect bounce of incident radiance to compute interreflections. The basic principle is to sample the points blocking direct light, or shedding indirect light, from coarser levels of the pyramid the farther away they are from a given receiver point. We unify the representation of visibility and indirect radiance at discrete azimuthal directions (i.e., as a function of a single elevation angle) using the concept of a casting set of visible points along this direction whose contributions are collected in the basis of normalized Legendre polynomials. This analytic representation is compact, requires no precomputation, and allows efficient integration to produce the spherical visibility and indirect radiance signals. Sub-sampling visibility and indirect radiance, while shading with full-resolution surface normals, further increases performance without introducing noticeable artifacts. Our method renders 512x512 height fields (>500K triangles) at 36Hz.
@article{DBLP:journals/cgf/NowrouzezahraiS09,
author = {Derek Nowrouzezahrai and
John M. Snyder},
title = {Fast Global Illumination on Dynamic Height Fields},
journal = {Comput. Graph. Forum},
volume = {28},
number = {4},
pages = {1131--1139},
year = {2009},
url = {https://doi.org/10.1111/j.1467-8659.2009.01490.x},
doi = {10.1111/J.1467-8659.2009.01490.X},
timestamp = {Mon, 13 May 2019 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/NowrouzezahraiS09.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Multi-objective shape segmentation and labeling.
Simari, P. D.; Nowrouzezahrai, D.; Kalogerakis, E.; and Singh, K.
Comput. Graph. Forum, 28(5): 1415–1425. 2009.
Paper (PDF)
Paper
doi
link
bibtex
abstract Shape segmentations designed for different applications show significant variation in the composition of their parts. In this paper, we introduce the segmentation and labeling of shape based on the optimization of multiple objectives that capture application-specific segmentation criteria. We present a number of efficient objective functions that capture useful shape adjectives (compact, flat, narrow, perpendicular, etc.) Segmentation descriptions within our framework combine multiple such objective functions with optional labels to define each part. The optimization problem is simplified by proposing weighted Voronoi partitioning as a compact and continuous parameterization of spatially embedded shape segmentations. Separation of spatially close but geodesically distant parts is made possible using multi-dimensional scaling (MDS) prior to Voronoi partitioning. Optimization begins with an initial segmentation found using the centroids of a k-means clustering of surface elements. This partition is automatically labeled to optimize heterogeneous part objectives and the Voronoi centers and their weights optimized using Generalized Pattern Search (GPS). We illustrate our framework using several diverse segmentation applications: consistent segmentations with semantic labels, bounding volume hierarchies for path tracing, and automatic rig and clothing transfer between animation characters.
@article{DBLP:journals/cgf/SimariNKS09,
author = {Patricio D. Simari and
Derek Nowrouzezahrai and
Evangelos Kalogerakis and
Karan Singh},
title = {Multi-objective shape segmentation and labeling},
journal = {Comput. Graph. Forum},
volume = {28},
number = {5},
pages = {1415--1425},
year = {2009},
url = {https://doi.org/10.1111/j.1467-8659.2009.01518.x},
doi = {10.1111/J.1467-8659.2009.01518.X},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/cgf/SimariNKS09.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Data-driven curvature for real-time line drawing of dynamic scenes.
Kalogerakis, E.; Nowrouzezahrai, D.; Simari, P. D.; McCrae, J.; Hertzmann, A.; and Singh, K.
ACM Trans. Graph., 28(1): 11:1–11:13. 2009.
Presented at SIGGRAPH
Paper (PDF)
Paper
doi
link
bibtex
abstract This paper presents a method for real-time line drawing of deforming objects. Object-space line drawing algorithms for many types of curves, including suggestive contours, highlights, ridges and valleys, rely on surface curvature and curvature derivatives. Unfortunately, these curvatures and their derivatives cannot be computed in real-time for animated, deforming objects. In a preprocessing step, our method learns the mapping from a low-dimensional set of animation parameters (e.g., joint angles) to surface curvatures for a deforming 3D mesh. The learned model can then accurately and efficiently predict curvatures and their derivatives, enabling real-time object-space rendering of suggestive contours and other such curves. This represents an order-of-magnitude speed-up over the fastest existing algorithm capable of estimating curvatures and their derivatives accurately enough for many different types of line drawings. The learned model can generalize to novel animation sequences, and is also very compact, typically requiring a few megabytes of storage at run-time. We demonstrate our method for various types of animated objects, including skeleton-based characters, cloth simulation and blend-shape facial animation, using a variety of non-photorealistic rendering styles.An important component of our system is the use of dimensionality reduction for differential mesh data.
@article{DBLP:journals/tog/KalogerakisNSMHS09,
author = {Evangelos Kalogerakis and
Derek Nowrouzezahrai and
Patricio D. Simari and
James McCrae and
Aaron Hertzmann and
Karan Singh},
title = {Data-driven curvature for real-time line drawing of dynamic scenes},
journal = {{ACM} Trans. Graph.},
volume = {28},
number = {1},
pages = {11:1--11:13},
year = {2009},
url = {https://doi.org/10.1145/1477926.1477937},
doi = {10.1145/1477926.1477937},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/journals/tog/KalogerakisNSMHS09.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2008
(3)
|
Fast Soft Self-Shadowing on Dynamic Height Fields.
Snyder, J. M.; and Nowrouzezahrai, D.
Comput. Graph. Forum, 27(4): 1275–1283. 2008.
Paper (PDF) Slides (PPT)
Paper
doi
link
bibtex
abstract We present a new, real-time method for rendering soft shadows from large light sources or lighting environments on dynamic height fields. The method first computes a horizon map for a set of azimuthal directions. To reduce sampling, we compute a multi-resolution pyramid on the height field. Coarser pyramid levels are indexed as the distance from caster to receiver increases. For every receiver point and every azimuthal direction, a smooth function of blocking angle in terms of log distance is reconstructed from a height difference sample at each pyramid level. This function's maximum approximates the horizon angle. We then sum visibility at each receiver point over wedges determined by successive pairs of horizon angles. Each wedge represents a linear transition in blocking angle over its azimuthal extent. It is precomputed in the order-4 spherical harmonic (SH) basis, for a canonical azimuthal origin and fixed extent, resulting in a 2D table. The SH triple product of 16D vectors representing lighting, total visibility, and diffuse reflectance then yields the soft-shadowed result. Two types of light sources are considered; both are distant and low-frequency. Environmental lights require visibility sampling around the complete 360 degree azimuth, while key lights sample visibility within a partial swath. Restricting the swath concentrates samples where the light comes from (e.g. 3 azimuthal directions vs. 16-32 for a full swath) and obtains sharper shadows. Our GPU implementation handles height fields up to 1024x1024 in real-time. The computation is simple, local, and parallel, with performance independent of geometric content.
@article{DBLP:journals/cgf/SnydreN08,
author = {John M. Snyder and
Derek Nowrouzezahrai},
title = {Fast Soft Self-Shadowing on Dynamic Height Fields},
journal = {Comput. Graph. Forum},
volume = {27},
number = {4},
pages = {1275--1283},
year = {2008},
url = {https://doi.org/10.1111/j.1467-8659.2008.01266.x},
doi = {10.1111/J.1467-8659.2008.01266.X},
timestamp = {Mon, 13 May 2019 01:00:00 +0200},
biburl = {https://dblp.org/rec/journals/cgf/SnydreN08.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Video browsing by direct manipulation.
Dragicevic, P.; Ramos, G. A.; Bibliowicz, J.; Nowrouzezahrai, D.; Balakrishnan, R.; and Singh, K.
In Czerwinski, M.; Lund, A. M.; and Tan, D. S., editor(s), Proceedings of the 2008 Conference on Human Factors in Computing Systems, CHI 2008, 2008, Florence, Italy, April 5-10, 2008, pages 237–246, 2008. ACM
Paper (PDF)
Paper
doi
link
bibtex
1 download
abstract We present a method for browsing videos by directly dragging their content. This method brings the benefits of direct manipulation to an activity typically mediated by widgets. We support this new type of interactivity by: 1) automatically extracting motion data from videos; and 2) a new technique called relative flow dragging that lets users control video playback by moving objects of interest along their visual trajectory. We show that this method can outperform the traditional seeker bar in video browsing tasks that focus on visual content rather than time.
@inproceedings{DBLP:conf/chi/DragicevicRBNBS08,
author = {Pierre Dragicevic and
Gonzalo A. Ramos and
Jacobo Bibliowicz and
Derek Nowrouzezahrai and
Ravin Balakrishnan and
Karan Singh},
editor = {Mary Czerwinski and
Arnold M. Lund and
Desney S. Tan},
title = {Video browsing by direct manipulation},
booktitle = {Proceedings of the 2008 Conference on Human Factors in Computing Systems,
{CHI} 2008, 2008, Florence, Italy, April 5-10, 2008},
pages = {237--246},
publisher = {{ACM}},
year = {2008},
url = {https://doi.org/10.1145/1357054.1357096},
doi = {10.1145/1357054.1357096},
timestamp = {Fri, 12 Mar 2021 15:27:48 +0100},
biburl = {https://dblp.org/rec/conf/chi/DragicevicRBNBS08.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Shadowed Relighting of Dynamic Geometry with 1D BRDFs.
Nowrouzezahrai, D.; Kalogerakis, E.; Simari, P. D.; and Fiume, E.
In Mania, K.; and Reinhard, E. R., editor(s), 29th Annual Conference of the European Association for Computer Graphics, Eurographics 2008 - Short Papers, Hersonissos, Crete, Greece, April 14-18, 2008, pages 115–118, 2008. Eurographics Association
Paper
doi
link
bibtex
3 downloads
abstract We present a method for synthesizing the dynamic self-occlusion of an articulating character in real-time (> 170Hz) while incorporating reflection effects from 1D BRDFs under dynamic lighting and view conditions. We introduce and derive a general operator form for convolving spherical harmonics (SH) occlusion vectors with arbitrary 1D BRDF kernels. This operator, coupled with a compact linear model for predicting SH occlusion over articulating meshes, segments the BRDF and visibility terms of the direct illumination integral. We illustrate our results on a thin-membrane translucency model and the normalized Phong BRDF.
@inproceedings{DBLP:conf/eurographics/NowrouzezahraiK08,
author = {Derek Nowrouzezahrai and
Evangelos Kalogerakis and
Patricio D. Simari and
Eugene Fiume},
editor = {Katerina Mania and
Eric R. Reinhard},
title = {Shadowed Relighting of Dynamic Geometry with 1D BRDFs},
booktitle = {29th Annual Conference of the European Association for Computer Graphics,
Eurographics 2008 - Short Papers, Hersonissos, Crete, Greece, April
14-18, 2008},
pages = {115--118},
publisher = {Eurographics Association},
year = {2008},
url = {https://doi.org/10.2312/egs.20081035},
doi = {10.2312/EGS.20081035},
timestamp = {Thu, 02 Jul 2020 14:52:18 +0200},
biburl = {https://dblp.org/rec/conf/eurographics/NowrouzezahraiK08.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2007
(4)
|
Compact and efficient generation of radiance transfer for dynamically articulated characters.
Nowrouzezahrai, D.; Simari, P. D.; Kalogerakis, E.; Singh, K.; and Fiume, E.
In Rohl, A. L., editor(s), Proceedings of the 5th International Conference on Computer Graphics and Interactive Techniques in Australasia and Southeast Asia 2007, Perth, Western Australia, 1-4 December 2007, pages 147–154, 2007. ACM
Paper (PDF) Slides (PDF)
Paper
doi
link
bibtex
1 download
abstract We present a data-driven technique for generating the precomputed radiance transfer vectors of an animated character as a function of its joint angles. We learn a linear model for generating real-time lighting effects on articulated characters while capturing soft self shadows caused by dynamic distant lighting. Indirect illumination can also be reproduced using our framework. Previous data-driven techniques have either restricted the type of lighting response (generating only ambient occlusion), the type of animated sequences (response functions to external forces) or have complicated runtime algorithms and incur non-trivial memory costs. We provide insights into the dimensionality reduction of the pose and coefficient spaces. Our model can be fit quickly as a preprocess, is very compact (~1MB) and runtime transfer vectors are generated using a simple algorithm in real-time (>100 Hz using a CPU-only implementation.) We can reproduce lighting effects on hundreds of trained poses using less memory than required to store a single mesh's PRT coefficients. Moreover, our model extrapolates to produce smooth, believable lighting results on novel poses and our method can be easily integrated into existing interactive content pipelines.
@inproceedings{DBLP:conf/graphite/NowrouzezahraiSKSF07,
author = {Derek Nowrouzezahrai and
Patricio D. Simari and
Evangelos Kalogerakis and
Karan Singh and
Eugene Fiume},
editor = {Andrew L. Rohl},
title = {Compact and efficient generation of radiance transfer for dynamically
articulated characters},
booktitle = {Proceedings of the 5th International Conference on Computer Graphics
and Interactive Techniques in Australasia and Southeast Asia 2007,
Perth, Western Australia, 1-4 December 2007},
pages = {147--154},
publisher = {{ACM}},
year = {2007},
url = {https://doi.org/10.1145/1321261.1321288},
doi = {10.1145/1321261.1321288},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/graphite/NowrouzezahraiSKSF07.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Eigentransport for efficient and accurate all-frequency relighting.
Nowrouzezahrai, D.; Simari, P. D.; Kalogerakis, E.; and Fiume, E.
In Rohl, A. L., editor(s), Proceedings of the 5th International Conference on Computer Graphics and Interactive Techniques in Australasia and Southeast Asia 2007, Perth, Western Australia, 1-4 December 2007, pages 163–169, 2007. ACM
Winner: Best Paper Award Winner!
Paper (PDF) Slides (PDF)
Paper
doi
link
bibtex
2 downloads
abstract We present a method for creating a geometry-dependent basis for diffuse precomputed radiance transfer. Unlike previous PRT bases, ours is derived from Principal Component Analysis of the sampled transport functions at each vertex, without relying on pre-projections to secondary bases, such as the Spherical Harmonics or Haar wavelets. It allows for efficient evaluation of shading, has low memory requirements and produces accurate results with few coefficients. We are able to capture all-frequency effects from both distant and near-field dynamic lighting in real-time and present a simple and efficient rotation scheme. Reconstruction of the final shading becomes a low-order dot product and is performed on the GPU.
@inproceedings{DBLP:conf/graphite/NowrouzezahraiSKF07,
author = {Derek Nowrouzezahrai and
Patricio D. Simari and
Evangelos Kalogerakis and
Eugene Fiume},
editor = {Andrew L. Rohl},
title = {Eigentransport for efficient and accurate all-frequency relighting},
booktitle = {Proceedings of the 5th International Conference on Computer Graphics
and Interactive Techniques in Australasia and Southeast Asia 2007,
Perth, Western Australia, 1-4 December 2007},
pages = {163--169},
publisher = {{ACM}},
year = {2007},
url = {https://doi.org/10.1145/1321261.1321290},
doi = {10.1145/1321261.1321290},
timestamp = {Mon, 03 Jan 2022 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/graphite/NowrouzezahraiSKF07.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Image-Based Proxy Accumulation for Real-Time Soft Global Illumination.
Sloan, P. J.; Govindaraju, N. K.; Nowrouzezahrai, D.; and Snyder, J. M.
In Alexa, M.; Gortler, S. J.; and Ju, T., editor(s), 15th Pacific Conference on Computer Graphics and Applications, PG 2007, Maui, HI, USA, October 29 - November 2, 2007, pages 97–105, 2007. IEEE Computer Society
Paper (PDF) Slides (PPTx)
Paper
doi
link
bibtex
abstract We present a new, general, and real-time technique for soft global illumination in low-frequency environmental lighting. It accumulates over relatively few spherical proxies which approximate the light blocking and re-radiating effect of dynamic geometry. Soft shadows are computed by accumulating log visibility vectors for each sphere proxy as seen by each receiver point. Inter-reflections are computed by accumulating vectors representing the proxy's unshadowed radiance when illuminated by the environment. Both vectors capture low-frequency directional dependence using the spherical harmonic basis. We also present a new proxy accumulation strategy that splats each proxy to receiver pixels in image space to collect its shadowing and indirect lighting contribution. Our soft GI rendering pipeline unifies direct and indirect soft effects with a simple accumulation strategy that maps entirely to the GPU and outperforms previous vertex-based methods.
@inproceedings{DBLP:conf/pg/SloanGNS07,
author = {Peter{-}Pike J. Sloan and
Naga K. Govindaraju and
Derek Nowrouzezahrai and
John M. Snyder},
editor = {Marc Alexa and
Steven J. Gortler and
Tao Ju},
title = {Image-Based Proxy Accumulation for Real-Time Soft Global Illumination},
booktitle = {15th Pacific Conference on Computer Graphics and Applications, {PG}
2007, Maui, HI, USA, October 29 - November 2, 2007},
pages = {97--105},
publisher = {{IEEE} Computer Society},
year = {2007},
url = {https://doi.org/10.1109/PG.2007.28},
doi = {10.1109/PG.2007.28},
timestamp = {Fri, 24 Mar 2023 00:00:00 +0100},
biburl = {https://dblp.org/rec/conf/pg/SloanGNS07.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
Robust statistical estimation of curvature on discretized surfaces.
Kalogerakis, E.; Simari, P. D.; Nowrouzezahrai, D.; and Singh, K.
In Belyaev, A. G.; and Garland, M., editor(s), Proceedings of the Fifth Eurographics Symposium on Geometry Processing, Barcelona, Spain, July 4-6, 2007, volume 257, of ACM International Conference Proceeding Series, pages 13–22, 2007. Eurographics Association
Paper (PDF)
Paper
doi
link
bibtex
abstract A robust statistics approach to curvature estimation on discretely sampled surfaces, namely polygon meshes and point clouds, is presented. The method exhibits accuracy, stability and consistency even for noisy, non-uniformly sampled surfaces with irregular configurations. Within an M-estimation framework, the algorithm is able to reject noise and structured outliers by sampling normal variations in an adaptively reweighted neighborhood around each point. The algorithm can be used to reliably derive higher order differential attributes and even correct noisy surface normals while preserving the fine features of the normal and curvature field. The approach is compared with state-of-the-art curvature estimation methods and shown to improve accuracy by up to an order of magnitude across ground truth test surfaces under varying tessellation densities and types as well as increasing degrees of noise. Finally, the benefits of a robust statistical estimation of curvature are illustrated by applying it to the popular applications of mesh segmentation and suggestive contour rendering.
@inproceedings{DBLP:conf/sgp/KalogerakisSNS07,
author = {Evangelos Kalogerakis and
Patricio D. Simari and
Derek Nowrouzezahrai and
Karan Singh},
editor = {Alexander G. Belyaev and
Michael Garland},
title = {Robust statistical estimation of curvature on discretized surfaces},
booktitle = {Proceedings of the Fifth Eurographics Symposium on Geometry Processing,
Barcelona, Spain, July 4-6, 2007},
series = {{ACM} International Conference Proceeding Series},
volume = {257},
pages = {13--22},
publisher = {Eurographics Association},
year = {2007},
url = {https://doi.org/10.2312/SGP/SGP07/013-022},
doi = {10.2312/SGP/SGP07/013-022},
timestamp = {Wed, 24 May 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/sgp/KalogerakisSNS07.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
2006
(2)
|
A controllable, fast and stable basis for vortex based smoke simulation.
Angelidis, A.; Neyret, F.; Singh, K.; and Nowrouzezahrai, D.
In O'Sullivan, C.; and Pighin, F. H., editor(s), Proceedings of the 2006 ACM SIGGRAPH/Eurographics Symposium on Computer Animation, SCA 2006, Vienna, Austria, September 2-4, 2006, pages 25–32, 2006. Eurographics Association
Paper
doi
link
bibtex
3 downloads
abstract We introduce a novel method for describing and controlling a 3D smoke simulation. Using harmonic analysis and principal component analysis, we define an underlying description of the fluid flow that is compact and meaningful to non-expert users. The motion of the smoke can be modified with high level tools, such as animated current curves, attractors and tornadoes. Our simulation is controllable, interactive and stable for arbitrarily long periods of time. The simulation's computational cost increases linearly in the number of motion samples and smoke particles. Our adaptive smoke particle representation conveniently incorporates the surface-like characteristics of real smoke.
@inproceedings{DBLP:conf/sca/AngelidisNSN06,
author = {Alexis Angelidis and
Fabrice Neyret and
Karan Singh and
Derek Nowrouzezahrai},
editor = {Carol O'Sullivan and
Fr{\'{e}}d{\'{e}}ric H. Pighin},
title = {A controllable, fast and stable basis for vortex based smoke simulation},
booktitle = {Proceedings of the 2006 {ACM} SIGGRAPH/Eurographics Symposium on Computer
Animation, {SCA} 2006, Vienna, Austria, September 2-4, 2006},
pages = {25--32},
publisher = {Eurographics Association},
year = {2006},
url = {https://doi.org/10.2312/SCA/SCA06/025-032},
doi = {10.2312/SCA/SCA06/025-032},
timestamp = {Fri, 26 May 2017 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/sca/AngelidisNSN06.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|
GPU-accelerated ray casting of node-based implicits.
Lessig, C.; Nowrouzezahrai, D.; and Singh, K.
In Finnegan, J. W.; and McGrath, M., editor(s), International Conference on Computer Graphics and Interactive Techniques, SIGGRAPH 2006, Boston, Massachusetts, USA, July 30 - August 3, 2006, Research Posters, pages 54, 2006. ACM
Paper
doi
link
bibtex
2 downloads
abstract We demonstrate that state-of-the-art GPUs are well suited for the visualization of node-based implicit surfaces that are the natural surface representation for Lagrangian simulations which are used for example for the simulation of fluids.
@inproceedings{DBLP:conf/siggraph/LessigNS06,
author = {Christian Lessig and
Derek Nowrouzezahrai and
Karan Singh},
editor = {John W. Finnegan and
Mike McGrath},
title = {GPU-accelerated ray casting of node-based implicits},
booktitle = {International Conference on Computer Graphics and Interactive Techniques,
{SIGGRAPH} 2006, Boston, Massachusetts, USA, July 30 - August 3, 2006,
Research Posters},
pages = {54},
publisher = {{ACM}},
year = {2006},
url = {https://doi.org/10.1145/1179622.1179684},
doi = {10.1145/1179622.1179684},
timestamp = {Fri, 12 Mar 2021 11:22:59 +0100},
biburl = {https://dblp.org/rec/conf/siggraph/LessigNS06.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
|
|