Proceedings ArticleDOI
Filling in scenes by propagating probabilities through layers and into appearance models
Brendan J. Frey
- Vol. 1, pp 185-192
TLDR
A Bayesian network is constructed that describes the occlusion process and iterative probability propagation is used to approximately recover the identities and positions of the objects in the scene in time that is linear in K and L.Abstract:
Inferring the identities and positions of multiple occluding objects in a noisy image is a difficult problem, even when the shapes and appearances of the allowable objects are known. Methods that detect and analyze shape features, occlusion boundaries and optical flow break down when the image is noisy. In situations where we know the boundaries and appearances of the allowable objects, a brute force method can be used to perform MAP inference. If there are K possible objects (including translations, etc.) in up to L layers, the number of possible configurations of the scene is K/sup L/, so exact inference is intractable for large numbers of objects and reasonably large numbers of layers. We construct a Bayesian network that describes the occlusion process and we use iterative probability propagation to approximately recover the identities and positions of the objects in the scene in time that is linear in K and L. Although iterative probability propagation is an approximate inference technique, it was recently used to obtain the world record in error-correcting decoding. Experiments show that when one explanation of the scene is most probable, the algorithm finds the solution. For a small problem, we show that as the number of iterations increases, iterative probability propagation performs better than a greedy technique and becomes closer to the exact MAP algorithm. Quite surprisingly, we also find that when the order of occlusion is ambiguous, the output of the algorithm may oscillate between plausible interpretations of the scene.read more
Citations
More filters
Journal ArticleDOI
Learning Low-Level Vision
TL;DR: A learning-based method for low-level vision problems—estimating scenes from images with Bayesian belief propagation, applied to the “super-resolution” problem (estimating high frequency details from a low-resolution image), showing good results.
Journal ArticleDOI
Robust online appearance models for visual tracking
TL;DR: A framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects to provide robustness in the face of image outliers, while adapting to natural changes in appearance such as those due to facial expressions or variations in 3D pose.
Journal ArticleDOI
Computer vision and pattern recognition
TL;DR: This Special Issue of International Journal of Computer Mathematics (IJCM) offers a venue to present innovative approaches in computer vision and pattern recognition, which have been changing the authors' everyday life dramatically over the last few years, and aims to provide readers with cutting-edge and topical information for their related research.
Proceedings ArticleDOI
Learning low-level vision
TL;DR: This work shows a learning-based method for low-level vision problems-estimating scenes from images with a Markov network, and applies VISTA to the "super-resolution" problem (estimating high frequency details from a low-resolution image), showing good results.
Proceedings ArticleDOI
Robust online appearance models for visual tracking
TL;DR: A framework for learning robust, adaptive appearance models to be used for motion-based tracking of natural objects and provides the ability to adapt to natural changes in appearance, such as those due to facial expressions or variations in 3D pose.
References
More filters
Journal ArticleDOI
Good error-correcting codes based on very sparse matrices
TL;DR: It is proved that sequences of codes exist which, when optimally decoded, achieve information rates up to the Shannon limit, and experimental results for binary-symmetric channels and Gaussian channels demonstrate that practical performance substantially better than that of standard convolutional and concatenated codes can be achieved.
Journal ArticleDOI
Local computations with probabilities on graphical structures and their application to expert systems
Journal ArticleDOI
Near optimum error correcting coding and decoding: turbo-codes
Claude Berrou,A. Glavieux +1 more
TL;DR: A new family of convolutional codes, nicknamed turbo-codes, built from a particular concatenation of two recursive systematic codes, linked together by nonuniform interleaving appears to be close to the theoretical limit predicted by Shannon.
Book ChapterDOI
A view of the EM algorithm that justifies incremental, sparse, and other variants
TL;DR: In this paper, an incremental variant of the EM algorithm is proposed, in which the distribution for only one of the unobserved variables is recalculated in each E step, which is shown empirically to give faster convergence in a mixture estimation problem.
Good error-correcting codes based on very sparse matrices (vol 45, pg 339, 1999)
TL;DR: It can be proved that, given an optimal decoder, Gallager's low density parity check codes asymptotically approach the Shannon limit.
Related Papers (5)
EigenTracking: Robust Matching and Tracking of Articulated Objects Using a View-Based Representation
Michael J. Black,Allan D. Jepson +1 more