Paper Info
Title
Multi-camera person tracking in a cluttered interaction environment
Abstract
Tracking the head and hand in real-time are important tasks for developing an intuitive interaction system. We present a system for robust probabilistic tracking that integrates face detection, face and hand color tracking and foot tracking in a uniform way by using particle filters. The advantages of different cues like motion, color and face detection are combined to yield robust 2D and 3D position estimates in spite of difficult varying lighting conditions and cluttered background. The system enables a user to navigate in the virtual scene by walking around and pointing towards objects by a simple hand gesture. The environment is a 3-sided CAVE with 1-sided stereo back projection.
Year
DOI
Venue
2005
10.1007/11556121_59
CAIP
Keywords
Field
DocType
different cue,face detection,3-sided cave,intuitive interaction system,cluttered background,robust probabilistic tracking,multi-camera person tracking,1-sided stereo,cluttered interaction environment,simple hand gesture,hand color tracking,foot tracking,real time
Facial recognition system,Computer vision,Pattern recognition,Motion detection,Computer science,Gesture,Particle filter,Image processing,Artificial intelligence,Probabilistic logic,Face detection,Motion estimation
Conference
Volume
ISSN
ISBN
3691
0302-9743
3-540-28969-0
Citations 
PageRank 
References 
0
0.34
8
Authors
2
Name
Order
Citations
PageRank
Daniel Grest112910.65
Reinhard Koch22038170.17