A framework for modeling the appearance of 3D articulated figures

A Survey of Computer Vision-Based Human Motion Capture

“…It is based on sampling the posterior distribution estimated in the previous frame and propagating these samples to form the posterior for the current frame. The method has shown to be a powerful alternative to the Kalman filter [112,130]. However, since it is nonparametric it requires a relatively large number of samples to ensure a fair maximum likelihood estimate of the current state.…”

“…This is done for several cameras, and by merging the results the 3D motion of the leg is estimated. In the work by Sidenbladh et al [130] a texture model of each limb is generated offline. They use a commercial motion capture system to obtain ground truth 3D pose data.…”

Pose Estimating the Human Arm Using Kinematics and the Sequential Monte Carlo Framework

“…This process is known as the analysis-by-synthesis (AbS) approach. The type of data used in the matching differ between systems, but usually it is: edges (Sminchisescu, 2002); (Wu et al, 2003), texture (Sidenbladh et al, 2000); (Ben-Arie et al, 2002), contours (Ong & Gong, 1999); (Delamarre & Faugeras, 2001), silhouettes (Moeslund & Granum, 2000); (Ogaki et al, 2001), or motion (Bregler & Malik, 1998); (Howe et al, 2000). The framework used to decide which synthesised data to match with the current image data also differ.…”

Tracking of Human Body Joints using Anthropometry

“…The Kalman filter has been used previously for human motion tracking [8,7], however, the use of the Kalman filter is limited by complex human dynamics. A strong alternative to the Kalman filter is the Condensation algorithm [4], employed by Ong in [6] and by Sidenbladh in [9]. In [1], Rehg modified the Condensation algorithm to overcome the problem of a large state space required for human motion tracking.…”