Vision, instruction, and action
Vision, instruction, and action
A Robust Algorithm for Video Based Human Motion Tracking
PCM '02 Proceedings of the Third IEEE Pacific Rim Conference on Multimedia: Advances in Multimedia Information Processing
An Introduction to the Kalman Filter
An Introduction to the Kalman Filter
3-D reconstruction of static human body shape from image sequence
Computer Vision and Image Understanding
Real-time and marker-free 3D motion capture for home entertainment oriented applications
ACCV'07 Proceedings of the 8th Asian conference on Computer vision - Volume Part I
Hi-index | 0.00 |
Most of the commercial motion capture techniques, be they marker-based or marker-free, are either expensive or require complex laboratory setup. These constraints limit their use for home entertainment or interactive gaming oriented applications. In this paper we present a real-time system for color marker-based human motion capture in a monocular image space for avatar animation. The proposed system requires a single webcam and cheap color markers for its operation resulting in low cost and user friendliness. Real-time efficiency is achieved by tracking through variable-sized window search based on a process noise-adaptive Kalman filter. The depth information is recovered with the aid of human anthropometric information in a perspective projection model. Finally experimental results of the system are presented as tested with a number of video sequences.