THEORY/MODELS

• IT-inspired divergences for shapes, images and graphs
• MDL, Maximum Entropy, Minimax Entropy and other IT principles
• Estimation of entropy and of other IT measures
• Information projection
• Information geometry
• IT kernels and kernel machines
• Integration with Bayesian/probabilistic models through IT
• The role of IT in the integration of bottom-up/top-down processes
• Alternative/complementary IT-like theories

APPLICATIONS

• Feature Extraction
• Early Vision, Attention and Context
• Image Segmentation
• Image Parsing and Recognition
• Image Registration
• Object Recognition and Categorization
• Cognitive Models for Perception
• Pattern Recognition with Grammars and Graphs
• Shape Matching and Learning
• Analysis of High-Dimensional Data
• Classifiers and I-Divergences
• Classification and Compression
• Active and on-Line Learning
• Learning Graphical Models
• Manifold Learning
• Unsupervised and Semi-Supervised Learning
• Clustering Ensembles and Consensus Methods
• Feature Transformation and Selection
• Quantification of Pattern Complexity
• Natural Images Statistics
• Domains: Biomedical Data, Remote Sensing, Multi-Media, 3D Shape Analysis, Robotics.