Markov Random Fields for Vision and Image Processing

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 developedalgorithms, as a means of making inferences about images. These inferences concern underlying imageand scene structure as well as solutions to such problems as image reconstruction, imagesegmentation, 3D vision, and object labeling. It offers key findings and state-of-the-art researchon 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 successfulapplications of MRFs, including segmentation, super-resolution, and image restoration, along with acomparison of various optimization methods; discusses advanced algorithmic topics; addresseslimitations of the strong locality assumptions in the MRFs discussed in earlier chapters; andshowcases applications that use MRFs in more complex ways, as components in bigger systems or withmultiterm energy functions. The book will be an essential guide to current research on thesepowerful mathematical tools.

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.