Unscented Kalman filtering for articulated human tracking
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Unscented Kalman filtering for articulated human tracking. / Boesen Lindbo Larsen, Anders; Hauberg, Søren; Pedersen, Kim Steenstrup.
Image Analysis: 17th Scandinavian Conference, SCIA 2011, Ystad, Sweden, May 2011. Proceedings. ed. / Anders Heyden; Fredrik Kahl. Springer, 2011. p. 228-237 (Lecture notes in computer science, Vol. 6688).Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Unscented Kalman filtering for articulated human tracking
AU - Boesen Lindbo Larsen, Anders
AU - Hauberg, Søren
AU - Pedersen, Kim Steenstrup
N1 - Conference code: 17
PY - 2011
Y1 - 2011
N2 - We present an articulated tracking system working with data from a single narrow baseline stereo camera. The use of stereo data allows for some depth disambiguation, a common issue in articulated tracking, which in turn yields likelihoods that are practically unimodal. While current state-of-the-art trackers utilize particle filters, our unimodal likelihood model allows us to use an unscented Kalman filter. This robust and efficient filter allows us to improve the quality of the tracker while using substantially fewer likelihood evaluations. The system is compared to one based on a particle filter with superior results. Tracking quality is measured by comparing with ground truth data from a marker-based motion capture system.
AB - We present an articulated tracking system working with data from a single narrow baseline stereo camera. The use of stereo data allows for some depth disambiguation, a common issue in articulated tracking, which in turn yields likelihoods that are practically unimodal. While current state-of-the-art trackers utilize particle filters, our unimodal likelihood model allows us to use an unscented Kalman filter. This robust and efficient filter allows us to improve the quality of the tracker while using substantially fewer likelihood evaluations. The system is compared to one based on a particle filter with superior results. Tracking quality is measured by comparing with ground truth data from a marker-based motion capture system.
U2 - 10.1007/978-3-642-21227-7_22
DO - 10.1007/978-3-642-21227-7_22
M3 - Article in proceedings
SN - 978-3-642-21226-0
T3 - Lecture notes in computer science
SP - 228
EP - 237
BT - Image Analysis
A2 - Heyden, Anders
A2 - Kahl, Fredrik
PB - Springer
Y2 - 23 May 2011 through 27 May 2011
ER -
ID: 170193916