Constraint Integration for Efficient Multiview Pose Estimation with Self-Occlusions

TitleConstraint Integration for Efficient Multiview Pose Estimation with Self-Occlusions
Publication TypeJournal Articles
Year of Publication2008
AuthorsGupta A, Mittal A, Davis LS
JournalPattern Analysis and Machine Intelligence, IEEE Transactions on
Pagination493 - 506
Date Published2008/03//
ISBN Number0162-8828
KeywordsAutomated;Posture;Reproducibility of Results;Sensitivity and Specificity;Video Recording;Whole Body Imaging;, automatic initialization;constraint integration;graphical structure;human pose tracking;kinematic constraint;likelihood measure;location probability distribution;multiview pose estimation;nonparametric belief propagation;optimization;pose configuration;se, Computer-Assisted;Imaging, Three-Dimensional;Pattern Recognition

Automatic initialization and tracking of human pose is an important task in visual surveillance. We present a part-based approach that incorporates a variety of constraints in a unified framework. These constraints include the kinematic constraints between parts that are physically connected to each other, the occlusion of one part by another, and the high correlation between the appearance of certain parts, such as the arms. The location probability distribution of each part is determined by evaluating appropriate likelihood measures. The graphical (nontree) structure representing the interdependencies between parts is utilized to "connect" such part distributions via nonparametric belief propagation. Methods are also developed to perform this optimization efficiently in the large space of pose configurations.