Skip to content. Skip to navigation
CIM Menus
 

Overcoming the Next Challenges in Vision and Language Research


Vicente Ordonez
University of Virginia

August 24, 2018 at  1:30 PM
McConnell Engineering Room 437

Abstract:

This talk will motivate and present three current challenges in vision and language research, and the current work of the vision and language group at the University of Virginia on addressing them: First, for building vision and language models that can demonstrably perform spatial reasoning as opposed to relying on textual statistics. Second, producing robust language and language-grounded models while minimizing their reliance on unwanted dataset biases or stereotypes. Third, building more general deep learning methods that can jointly model images and text for integrated reasoning tasks such as conditional inference, including the use of feedback-loops.

Bio:

Vicente Ordonez is Assistant Professor of Computer Science at the University of Virginia in the United States since the Fall 2016. He was previously a Visiting Research Fellow at the Allen Institute for Artificial Intelligence, and obtained a PhD in Computer Science at the University of North Carolina at Chapel Hill In 2015. His research has been at the intersection of vision and language, and deep learning for visual recognition and large scale image analysis. He has been a recipient of the Marr Prize at the International Conference in Computer Vision (ICCV) in 2013, and a Best Paper Award at the conference on Empirical Methods in Natural Language Processing (EMNLP) in 2017. He has received faculty research awards from Google and IBM.