Complex Visual Motion Analysis via Probabilistic Variational Methods : A Distributed and Collaborative Approach

ISBN-13: 9783639117103 / Angielski / Miękka / 2009 / 140 str.

Gang Hua

Opis

The explosive growth of the internet have accumulated gigantic amount of visual data. Management and retrieval of relevant information from it become a more and more difficult task. Moreover, as security becomes an important issue, there are numerous surveillance cameras all around the world which accumulate terabytes of video data every day. How to intelligently analyze these video data becomes a big challenge. These largely explain the boom of the research on complex motion analysis, the understanding of which would be essential for higher level tasks such as visual recognition and content understanding. This book provides a general framework on complex motion analysis based on probabilistic variational method. It provides a distributed computational paradigm, since each motion component is analyzed individually, with the constraints from the other components being reinforced. It is also a collaborative approach, since it naturally integrates multiple information cues to obtain the optimal solution.

O autorze

Dr. Gang Hua is a Scientist in computer vision, pattern
recognition, and machine learning in Microsoft Live labs
Research. He obtained his Ph.D. degree in Electrical and
Computer Engineering from Northwestern University in June 2006.
He has authored more than 30 peer reviewed papers. He also holds
1 US patent and has 15+ patents pending.