The articulated body models used to represent human motion typically have many degrees of freedom, usually expressed as joint angles that are highly correlated. T...
Andrea Fossati (EPFL), Mathieu Salzmann (Universit...
We propose a joint representation and classification framework that achieves the dual goal of finding the most discriminative sparse overcomplete encoding and optimal classifier p...
We show that the automatically induced latent variable grammars of Petrov et al. (2006) vary widely in their underlying representations, depending on their EM initialization point...
Top-down visual saliency facilities object localization by providing a discriminative representation of target objects and a probability map for reducing the search space. In this...
Learning graphical models with hidden variables can offer semantic insights to complex data and lead to salient structured predictors without relying on expensive, sometime unatta...