• Publications
  • Influence
Dynamic Textures
A characterization of dynamic textures that poses the problems of modeling, learning, recognizing and synthesizing dynamic textures on a firm analytical footing and experimental evidence that, within the framework, even low-dimensional models can capture very complex visual phenomena is presented. Expand
Meta-Learning With Differentiable Convex Optimization
The objective is to learn feature embeddings that generalize well under a linear classification rule for novel categories and this work exploits two properties of linear classifiers: implicit differentiation of the optimality conditions of the convex problem and the dual formulation of the optimization problem. Expand
An Invitation to 3-D Vision: From Images to Geometric Models
This book introduces the geometry of3-D vision, that is, the reconstruction of 3-D models of objects from a collection of 2-D images, and develops practical reconstruction algorithms and discusses possible extensions of the theory. Expand
Quick Shift and Kernel Methods for Mode Seeking
We show that the complexity of the recently introduced medoid-shift algorithm in clustering N points is O(N 2), with a small constant, if the underlying distance is Euclidean. This makes medoid shiftExpand
Entropy-SGD: Biasing Gradient Descent Into Wide Valleys
This paper proposes a new optimization algorithm called Entropy-SGD for training deep neural networks that is motivated by the local geometry of the energy landscape and compares favorably to state-of-the-art techniques in terms of generalization error and training time. Expand
Dynamic texture recognition
This work poses the problem of recognizing and classifying dynamic textures in the space of dynamical systems where each dynamic texture is uniquely represented and examines three different distances in thespace of autoregressive models and assess their power. Expand
Class segmentation and object localization with superpixel neighborhoods
A method to identify and localize object classes in images by constructing a classifier on the histogram of local features found in each superpixel using superpixels as the basic unit of a class segmentation or pixel localization scheme. Expand
Information Dropout: Learning Optimal Representations Through Noisy Computation
It is proved that Information Dropout achieves a comparable or better generalization performance than binary dropout, especially on smaller models, since it can automatically adapt the noise to the structure of the network, as well as to the test sample. Expand
Kernel Density Estimation and Intrinsic Alignment for Shape Priors in Level Set Segmentation
This paper proposes shape dissimilarity measures on the space of level set functions which are analytically invariant under the action of certain transformation groups, and proposes a statistical shape prior which allows to accurately encode multiple fairly distinct training shapes. Expand