Abstract
Particle filters allow for visual trackers with nonlinear measurements. In this paper we consider three different non-linear visual measurement cues, based on object detection, foreground segmentation and colour matching. Novel ways to obtain robust measurement likelihoods under a unified representation scheme are discussed, followed by a likelihood combination scheme for fusion. The resulting single and multi-cue particle filter trackers are compared in the scope of face tracking.
Original language | English |
---|---|
Title of host publication | 2013 18th International Conference on Digital Signal Processing, DSP 2013 |
Number of pages | 6 |
Publisher | IEEE Press |
Publication date | 6 Dec 2013 |
Pages | 1-6 |
Article number | 6622722 |
ISBN (Print) | 978-146735805-7 |
ISBN (Electronic) | 978-1-4673-5807-1 |
DOIs | |
Publication status | Published - 6 Dec 2013 |
Event | International conference on Digital Signal Processing - Fira, Greece Duration: 1 Jul 2013 → 3 Jul 2013 Conference number: 18 http://dsp2013.dspconferences.org/ |
Conference
Conference | International conference on Digital Signal Processing |
---|---|
Number | 18 |
Country/Territory | Greece |
City | Fira |
Period | 01/07/2013 → 03/07/2013 |
Internet address |
Series | International Conference on Digital Signal Processing proceedings |
---|---|
ISSN | 1546-1874 |
Keywords
- Face tracking
- Fusion
- Likelihood function
- Particle filters
- Visual measurements