Markov Random Fields for Vision and Image Processing

by Blake, Andrew; Kohli, Pushmeet; Rother, Carsten

• ISBN13: 9780262015776
• ISBN10: 0262015773
• Format: Hardcover
• Copyright: 2011-07-22
• Publisher: The MIT Press

Summary

This volume demonstrates the power of the Markov random field (MRF) in vision, treating the MRF both as a tool for modeling image data and, utilizing recently developed algorithms, as a means of making inferences about images. These inferences concern underlying image and scene structure as well as solutions to such problems as image reconstruction, image segmentation, 3D vision, and object labeling. It offers key findings and state-of-the-art research on both algorithms and applications. After an introduction to the fundamental concepts used in MRFs, the book reviews some of the main algorithms for performing inference with MRFs; presents successful applications of MRFs, including segmentation, super-resolution, and image restoration, along with a comparison of various optimization methods; discusses advanced algorithmic topics; addresses limitations of the strong locality assumptions in the MRFs discussed in earlier chapters; and showcases applications that use MRFs in more complex ways, as components in bigger systems or with multiterm energy functions. The book will be an essential guide to current research on these powerful mathematical tools.

Author Biography

Andrew Blake is Managing Director of Microsoft Research Cambridge (UK), where he has led the Computer Vision Research Group since 1999. Pushmeet Kohli and Carsten Rother are researchers in the Computer Vision Group at Microsoft Research Cambridge.

Table of Contents

Introduction to Markov Random Fields	p. 1
Algorithms for Inference of MAP Estimates for MRFs	p. 29
Basic Graph Cut Algorithms	p. 31
Optimizing Multilabel MRFs Using Move-Making Algorithms
Optimizing Multilabel MRFs with Convex and Truncated Convex Priors
Loopy Belief Propagation, Mean Field Theory, and Bethe Approximations	p. 77
Linear Programming and Variants of Belief Propagation	p. 95
Applications of MRFs, Including Segmentation	p. 109
Interactive Foreground Extraction: Using Graph Cut	p. 111
Continuous-Valued MRF for Image Segmentation	p. 127
Bilayer Segmentation of Video	p. 143
MRFs for Superresolution and Texture Synthesis	p. 155
A Comparative Study of Energy Minimization Methods for MRFs	p. 167
Further Topics: Inference, Parameter Learning, and Continuous Models	p. 183
Convex Relaxation Techniques for Segmentation, Stereo, and Multiview Reconstruction	p. 185
Learning Parameters in Continuous-Valued Markov Random Fields	p. 201
Message Passing with Continuous Latent Variables	p. 215
Learning Large-Margin Random Fields Using Graph Cuts	p. 233
Analyzing Convex Relaxations for MAP Estimation	p. 249
MAP Inference by Fast Primal-Dual Linear Programming	p. 263
Fusion-Move Optimization for MRFs with an Extensive Label Space	p. 281
Higher-Order MRFs and Global Constraints	p. 295
Field of Experts	p. 297
Enforcing Label Consistency Using Higher-Order Potentials	p. 311
Exact Optimization for Markov Random Fields with Nonlocal Parameters	p. 329
Graph Cut-Based Image Segmentation with Connectivity Priors	p. 347
Advanced Applications of MRFs	p. 363
Symmetric Stereo Matching for Occlusion Handling	p. 365
Steerable Random Fields for Image Restoration	p. 377
Markov Random Fields for Object Detection	p. 389
SIFT Flow: Dense Correspondence across Scenes and Its Applications	p. 405
Unwrap Mosaics: A Model for Deformable Surfaces in Video	p. 419
Bibliography	p. 433
Contributors	p. 457
Index	p. 459
Table of Contents provided by Ingram. All Rights Reserved.