Synopses & Reviews
Techniques of vision-based motion analysis aim to detect, track, identify, and generally understand the behavior of objects in image sequences. With the growth of video data in a wide range of applications from visual surveillance to human-machine interfaces, the ability to automatically analyze and understand object motions from video footage is of increasing importance. Among the latest developments in this field is the application of statistical machine learning algorithms for object tracking, activity modeling, and recognition. Developed from expert contributions to the first and second International Workshop on Machine Learning for Vision-Based Motion Analysis, this important text/reference highlights the latest algorithms and systems for robust and effective vision-based motion understanding from a machine learning perspective. Highlighting the benefits of collaboration between the communities of object motion understanding and machine learning, the book discusses the most active forefronts of research, including current challenges and potential future directions. Topics and features: Provides a comprehensive review of the latest developments in vision-based motion analysis, presenting numerous case studies on state-of-the-art learning algorithmsExamines algorithms for clustering and segmentation, and manifold learning for dynamical modelsDescribes the theory behind mixed-state statistical models, with a focus on mixed-state Markov models that take into account spatial and temporal interactionDiscusses object tracking in surveillance image streams, discriminative multiple target tracking, and guidewire tracking in fluoroscopyExplores issues of modeling for saliency detection, human gait modeling, modeling of extremely crowded scenes, and behavior modeling from video surveillance dataInvestigates methods for automatic recognition of gestures in Sign Language, and human action recognition from small training setsResearchers, professional engineers, and graduate students in computer vision, pattern recognition and machine learning, will all find this text an accessible survey of machine learning techniques for vision-based motion analysis. The book will also be of interest to all who work with specific vision applications, such as surveillance, sport event analysis, healthcare, video conferencing, and motion video indexing and retrieval. Dr. Liang Wang is a lecturer at the Department of Computer Science at the University of Bath, UK, and is also affiliated to the National Laboratory of Pattern Recognition in Beijing, China. Dr. Guoying Zhao is an adjunct professor at the Department of Electrical and Information Engineering at the University of Oulu, Finland. Dr. Li Cheng is a research scientist at the Agency for Science, Technology and Research (A*STAR), Singapore. Dr. Matti Pietikäinen is Professor of Information Technology at the Department of Electrical and Information Engineering at the University of Oulu, Finland.
From the reviews: "The successes of the First and Second International Workshops on Machine Learning for Vision-Based Motion Analysis, which were held in 2008 and 2009, prompted this book. The book consists of four parts, and each part includes a number of freestanding chapters. ... This book provides a comprehensive introduction to machine learning for vision-based motion analysis. I would recommend it to students and researchers who are interested in learning about the topic." (J. P. E. Hodgson, ACM Computing Reviews, June, 2011)
Vision-based motion analysis aims to detect, track and identify visual objects, and more generally, to understand their behaviors, from video sequences. This exciting research area has received growing interest in recent years due to a wide spectrum of proposing applications such as visual surveillance, human-machine interface, virtual reality, and motion analysis. Statistical machine learning algorithms have been recently successfully applied to address various challenging problems involved in this area. This book is intended to be a useful reference book that contains an excellent collection of theoretical and technical chapters written by the authors who are worldwide-recognized researchers on various aspects of motion analysis using machine learning methods.
* Uniquely focuses on modeling, analyzing, and understanding different types of motion from a machine learning perspective.
* Systematically reviews this research area and provides promising future research directions
* Provides the reader with a clear picture of the most active research forefronts and discussions of future directions, which different levels of researchers will find useful for guiding their future research
* Includes chapters expanded from invited talks, orals and posters from the 1st International Workshop on Machine Learning for Vision-based Motion Analysis (MLVMA'08), in conjunction with the European Conference on Computer Vision 2008 (ECCV'08).
Based on contributions to the International Workshop on Machine Learning for Vision-Based Motion Analysis, this volume highlights the latest algorithms and systems for robust and effective vision-based motion understanding.
Table of Contents
Part I: Manifold Learning and Clustering/Segmentation Practical Algorithms of Spectral Clustering: Toward Large-Scale Vision-Based Motion Analysis Tomoya Sakai, and Atsushi Imiya Riemannian Manifold Clustering and Dimensionality Reduction for Vision-based Analysis Alvina Goh Manifold Learning for Multi-dimensional Auto-regressive Dynamical Models Fabio Cuzzolin Part II: Tracking Mixed-state Markov Models in Image Motion Analysis Tomás Crivelli, Patrick Bouthemy, Bruno Cernuschi Frías, and Jian-feng Yao Learning to Detect Event Sequences in Surveillance Streams at Very Low Frame Rate Paolo Lombardi, and Cristina Versino Discriminative Multiple Target Tracking Xiaoyu Wang, Gang Hua, and Tony X. Han A Framework of Wire Tracking in Image Guided Interventions Peng Wang, Andreas Meyer, Terrence Chen, Shaohua K. Zhou, and Dorin Comaniciu Part III: Motion Analysis and Behavior Modeling An Integrated Approach to Visual Attention Modeling for Saliency Detection in Videos Sunaad Nataraju, Vineeth Balasubramanian, and Sethuraman Panchanathan Video-based Human Motion Estimation by Part-whole Gait Manifold Learning Guoliang Fan, and Xin Zhang Spatio-temporal Motion Pattern Models of Extremely Crowded Scenes Louis Kratz and Ko Nishino Learning Behavioral Patterns of Time Series for Video-surveillance Nicoletta Noceti, Matteo Santoro, and Francesca Odone Part IV: Gesture and Action Recognition Recognition of Spatiotemporal Gestures in Sign Language using Gesture Threshold HMMs Daniel Kelly, John Mc Donald and Charles Markham Learning Transferable Distance Functions for Human Action Recognition Weilong Yang, YangWang, and Greg Mori