Abstract
Visual tracking of human body movement is a key technology in a number of areas. In this paper we present a 2-D model-based method of human body tracking from a monocular video sequence. Morris & Rehg putforward a 2-D Scaled Prismatic Model (SPM) for figure registration which has far fewer singularity problems than 3-D models. Here we extend it in a 2-D cardboard human body model with additional one DOF of width change. We set up a mixture motion model for body movements and then solve body motion parameters using EM in a statistical framework, where the model-based kinematic constraints are incorporated in a linear form. Tracking results from real video sequences are encouraging.
Original language | English (US) |
---|---|
Pages (from-to) | 552-555 |
Number of pages | 4 |
Journal | Proceedings - International Conference on Pattern Recognition |
Volume | 16 |
Issue number | 1 |
State | Published - 2002 |
Externally published | Yes |
Keywords
- Cardboard body model
- Expectation-maximization
- Human tracking
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition