On-line selection of discriminative tracking features
- 1 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 346-352 vol.1
- https://doi.org/10.1109/iccv.2003.1238365
Abstract
We present a method for evaluating multiple feature spaces while tracking, and for adjusting the set of features used to improve tracking performance. Our hypothesis is that the features that best discriminate between object and background are also best for tracking the object. We develop an online feature selection mechanism based on the two-class variance ratio measure, applied to log likelihood distributions computed with respect to a given feature from samples of object and background pixels. This feature selection mechanism is embedded in a tracking system that adaptively selects the top-ranked discriminative features for tracking. Examples are presented to illustrate how the method adapts to changing appearances of both tracked object and scene background.Keywords
This publication has 8 references indexed in Scilit:
- Facial asymmetry quantification for expression invariant human identificationComputer Vision and Image Understanding, 2003
- Adaptive color space switching for face tracking in multi-colored lighting environmentsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Adaptive background mixture models for real-time trackingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Comparison of five color models in skin pixel classificationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Real-time tracking of non-rigid objects using mean shiftPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A unified approach to moving object detection in 2D and 3D scenesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1998
- CONDENSATION—Conditional Density Propagation for Visual TrackingInternational Journal of Computer Vision, 1998
- Note on optimal selection of independent binary-valued features for pattern recognition (Corresp.)IEEE Transactions on Information Theory, 1971