This work proposes a learning method for deep architectures that takes advantage of sequential data, in particular from the temporal coherence that naturally exists in unlabeled v...
We develop a framework for 3?D shape and motion recovery of articulated deformable objects. We propose a formalism that incorporates the use of implicit surfaces into earlier robo...
SynthesizedTextures represent continuous-tone images with a small set of vectors and color profiles. This representation is not only efficient in terms of compression ratio but al...
In this paper we propose a new method for human action categorization by using an effective combination of a new 3D gradient descriptor with an optic flow descriptor, to represent...