CUED Publications database

Fast upper body joint tracking using Kinect pose priors

Burke, M and Lasenby, J (2014) Fast upper body joint tracking using Kinect pose priors. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8563 L. pp. 94-105. ISSN 0302-9743

Full text not available from this repository.


Traditional approaches to upper body pose estimation using monocular vision rely on complex body models and a large variety of geometric constraints. We argue that this is not ideal and instead attempt to incorporate these constraints through priors obtained directly from training data, by fitting a Gaussian mixture model to a large dataset of recorded human body poses, tracked using a Kinect sensor. We combine this information with a random walk transition model to obtain an upper body model that can be viewed as a mixture of discrete Ornstein-Uhlenbeck processes, in that states behave as random walks, but drift towards a set of typically observed poses. The suggested model is designed with analytical tractability in mind and we show that the pose tracking can be Rao-Blackwellised using the mixture Kalman filter, allowing for computational efficiency while still incorporating bio-mechanical properties of the upper body. © 2014 Springer International Publishing.

Item Type: Article
Divisions: Div F > Signal Processing and Communications
Depositing User: Cron Job
Date Deposited: 17 Jul 2017 19:16
Last Modified: 26 Aug 2021 07:08
DOI: 10.1007/978-3-319-08849-5_10