Using an Adaptive VAR Model for Motion Prediction in 3D Hand Tracking

D. Chik, J. Trumpf, and N. N. Schraudolph. Using an Adaptive VAR Model for Motion Prediction in 3D Hand Tracking. In 8^th Intl. Conf. Automatic Face & Gesture Recognition (FG), IEEE, Amsterdam, Netherlands, 2008.

Download

1.4MB	818.1kB

Abstract

A robust VAR-based (vector autoregressive) model is introduced for motion prediction in 3D hand tracking. This dynamic VAR motion model is learned in an online manner. The kinematic structure of the hand is accounted for in the form of constraints when solving for the parameters of the VAR model. Also integrated into the motion prediction model are adaptive weights that are optimised according to the reliability of past predictions. Experiments on synthetic and real video sequences show a substantial improvement in tracking performance when the robust VAR motion model is used. In fact, utilising the robust VAR model allows the tracker to handle fast out-of-plane hand movement with severe self-occlusion.

BibTeX Entry

@inproceedings{ChiTruSch08,
     author = {Desmond Chik and Jochen Trumpf and Nicol N. Schraudolph},
      title = {\href{http://nic.schraudolph.org/pubs/ChiTruSch08.pdf}{
               Using an Adaptive {VAR} Model for Motion Prediction
               in 3D Hand Tracking}},
  booktitle = {8$^{th}$ Intl.\ Conf.\ Automatic
               Face \& Gesture Recognition (FG)},
    address = {Amsterdam, Netherlands},
  publisher = {IEEE},
       year =  2008,
   b2h_type = {Other},
  b2h_topic = {>Stochastic Meta-Descent, Computer Vision},
  abstract = {
    A robust VAR-based (vector autoregressive) model is introduced
    for motion prediction in 3D hand tracking. This dynamic VAR
    motion model is learned in an online manner. The kinematic
    structure of the hand is accounted for in the form of constraints
    when solving for the parameters of the VAR model. Also integrated
    into the motion prediction model are adaptive weights that are
    optimised according to the reliability of past predictions.
    Experiments on synthetic and real video sequences show a
    substantial improvement in tracking performance when the robust
    VAR motion model is used. In fact, utilising the robust VAR
    model allows the tracker to handle fast out-of-plane hand
    movement with severe self-occlusion.
}}