Statistical and Geometrical Approaches to Visual Motion Analysis

This book constitutes the thoroughly refereed post-conference proceedings of the International Dagstuhl-Seminar on Statistical and Geometrical Approaches to Visual Motion Analysis, held in Dagstuhl Castle, Germany, in July 2008.

The workshop focused on critical aspects of motion analysis, including motion segmentation and the modeling of motion patterns. The aim was to gather researchers who are experts in the different motion tasks and in the different techniques used; also involved were experts in the study of human and primate vision.

The 15 revised full papers presented were carefully reviewed and selected from or initiated by the lectures given at the workshop. The papers are organized in topical sections on optical flow and extensions, human motion modeling, biological and statistical approaches, alternative approaches to motion analysis.

Contenu

Optical Flow and Extensions.- Discrete-Continuous Optimization for Optical Flow Estimation.- An Improved Algorithm for TV-L 1 Optical Flow.- An Evaluation Approach for Scene Flow with Decoupled Motion and Position.- An Affine Optical Flow Model for Dynamic Surface Reconstruction.- Deinterlacing with Motion-Compensated Anisotropic Diffusion.- Human Motion Modeling.- Real-Time Synthesis of Body Movements Based on Learned Primitives.- 2D Human Pose Estimation in TV Shows.- Recognition and Synthesis of Human Movements by Parametric HMMs.- Recognizing Human Actions by Their Pose.- Biological and Statistical Approaches.- View-Based Approaches to Spatial Representation in Human Vision.- Combination of Geometrical and Statistical Methods for Visual Navigation of Autonomous Robots.- Motion Integration Using Competitive Priors.- Alternative Approaches to Motion Analysis.- Derivation of Motion Characteristics Using Affine Shape Adaptation for Moving Blobs.- Comparison of Point and Line Features and Their Combination for Rigid Body Motion Estimation.- The Conformal Monogenic Signal of Image Sequences.