Information theory has proved to be effective for solving many computer vision and pattern recognition (CVPR) problems (such as image matching, clustering and segmentation, saliency detection, feature selection, optimal classifier design and many others). Nowadays, researchers are widely bringing information theory elements to the CVPR arena. Among these elements there are measures (entropy, mutual information…), principles (maximum entropy, minimax entropy…) and theories (rate distortion theory, method of types…). This book explores and introduces the latter elements through an incremental complexity approach at the same time where CVPR problems are formulated and the most representative algorithms are presented. Interesting connections between information theory principles when applied to different problems are highlighted, seeking a comprehensive research roadmap. The result is a novel tool both for CVPR and machine learning researchers, and contributes to a cross-fertilization of both areas.

1 Introduction -- 1.1 Measures, Principles, Theories, and More -- 1.2 Detailed Organization of the Book -- 1.3 The ITinCVPR Roadmap -- 2 Interest Points, Edges, and Contour Grouping -- 2.1 Introduction -- 2.2 Entropy and Interest Points -- 2.2.1 Kadir and Brady Scale Saliency Detector -- 2.2.2 Point Filtering by Entropy Analysis Through Scale Space -- 2.2.3 Chernoff Information and Optimal Filtering -- 2.2.4 Bayesian Filtering of the Scale Saliency Feature Extractor: The Algorithm -- 2.3 Information Theory as Evaluation Tool: The Statistical Edge Detection Case -- 2.3.1 Statistical Edge Detection -- 2.3.2 Edge Localization -- 2.4 Finding Contours Among Clutter -- 2.4.1 Problem Statement -- 2.4.2 AddXMLRootTags.pl Road Tracking -- 2.4.3 AddXMLRootTags.pl Convergence Proof -- 2.5 Junction Detection and Grouping -- 2.5.1 Junction Detection -- 2.5.2 Connecting and Filtering Junctions -- Problems -- 2.6 Key References -- 3 Contour and Region-Based Image Segmentation -- 3.1 Introduction -- 3.2 Discriminative Segmentation with JensenShannon Divergence -- 3.2.1 The Active Polygons Functional -- 3.2.2 JensenShannon Divergence and the Speed Function -- 3.3 MDL in Contour-Based Segmentation -- 3.3.1 B-Spline Parameterization of Contours -- 3.3.2 MDL for B-Spline Parameterization -- 3.3.3 MDL Contour-based Segmentation -- 3.4 Model Order Selection in Region-Based Segmentation -- 3.4.1 Jump-Diffusion for Optimal Segmentation -- 3.4.2 Speeding-up the Jump-Diffusion Process -- 3.4.3 K-adventurers Algorithm -- 3.5 Model-Based Segmentation Exploiting The Maximum Entropy Principle -- 3.5.1 Maximum Entropy and Markov Random Fields -- 3.5.2 Efficient Learning with Belief Propagation -- 3.6 Integrating Segmentation, Detection and Recognition -- 3.6.1 Image Parsing -- 3.6.2 The Data-Driven Generative Model -- 3.6.3 The Power of Discriminative Processes -- 3.6.4 The Usefulness of Combining Generative and Discriminative -- Problems -- 3.7 Key References -- 4 Registration, Matching, and Recognition -- 4.1 Introduction -- 4.2 Image Alignment and Mutual Information -- 4.2.1 Alignment and Image Statistics -- 4.2.2 Entropy Estimation with Parzens Windows -- 4.2.3 The EMMA Algorithm -- 4.2.4 Solving the Histogram-Binning Problem -- 4.3 Alternative Metrics for Image Alignment -- 4.3.1 Normalizing Mutual Information -- 4.3.2 Conditional Entropies -- 4.3.3 Extension to the Multimodal Case -- 4.3.4 Affine Alignment of Multiple Images -- 4.3.5 The R233;nyi Entropy -- 4.3.6 R233;nyis Entropy and Entropic Spanning Graphs -- 4.3.7 The JensenR233;nyi Divergence and Its Applications -- 4.3.8 Other Measures Related to R233;nyi Entropy -- 4.3.9 Experimental Results -- 4.4 Deformable Matching with Jensen Divergence and Fisher Information -- 4.4.1 The Distributional Shape Model -- 4.4.2 Multiple Registration and JensenShannon Divergence -- 4.4.3 Information Geometry and FisherRao Information -- 4.4.4 Dynamics of the Fisher Information Metric -- 4.5 Structural Learning with MDL -- 4.5.1 The Usefulness of Shock Trees -- 4.5.2 A Generative Tree Model Based on Mixtures -- 4.5.3 Learning the Mixture -- 4.5.4 Tree Edit-Distance and MDL -- Problems -- 4.6 Key References -- 5 Image and Pattern Clustering -- 5.1 Introduction -- 5.2 Gaussian Mixtures and Model Selection -- 5.2.1 Gaussian Mixtu

pbk.