Caltech Library System

About | Browse | Search | Caltech Student Instructions

Type of Document	Master's Thesis
Author	Fanti, Claudio
URN	etd-05282004-144523
Persistent URL	http://resolver.caltech.edu/CaltechETD:etd-05282004-144523
Title	An improved scheme for detection and labeling in Johansson displays
Degree	Master of Science
Option	Computer Science
Advisory Committee	Advisor Name Title Pietro Perona Committee Chair
Keywords	moving light display Johansson display conditional independence expectation maximization graphical models Loopy Belief Propagation
Date of Defense	2004-06-11
Availability	unrestricted
Abstract Consider a number of moving points, where each point is attached to a joint of the human body and projected onto an image plane. Johannson showed that humans can effortlessly detect and recognize the presence of other humans from such displays. This is true even when some of the body points are missing (e.g. because of occlusion) and unrelated clutter points are added to the display. We are interested in replicating this ability in a machine. To this end, we present a labeling and detection scheme in a probabilistic framework. Our method is based on representing the joint probability density of positions and velocities of body points with a graphical model, and using Loopy Belief Propagation to calculate a likely interpretation of the scene. Furthermore, we introduce a global variable representing the body's centroid. Experiments on one motion-captured sequence suggest that our scheme improves on the accuracy of a previous approach based on triangulated graphical models, especially when very few parts are visble. The improvement is due both to the more general graph structure we use and, more significantly, to the introduction of the centroid variable.
Files	Filename       Size       Approximate Download Time (Hours:Minutes:Seconds)     28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access    MSThesis.pdf 291.87 Kb 00:01:21 00:00:41 00:00:36 00:00:18 00:00:01