This work considers an approach for artificially enhancing the richness and level of detail of graphical scenes. In particular, we examine a method for automatically generating high-resolution novel curves from manually sketched drawings of those curves. The essential idea is to augment the hand-drawn curves using prior knowledge to produce a more elaborated picture. Our method uses multi-scale analysis of a class of training data to capture statistical properties of the set. These properties are then conditioned at a coarse scale by the hand-drawn curve to {\it steers} the synthesis according to the overall shape of the curve. Given an approximation sketch, the algorithm generates the most likely scene by propagating probabilities over a Markov Chain. Users without artistic capabilities can then describe scenes in a more natural way to build impressive graphics.