Kernel-based object tracking

@article{citeulike:802362, abstract = {A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed. The feature histogram-based target representations are regularized by spatial masking with an isotropic kernel. The masking induces spatially-smooth similarity functions suitable for gradient-based optimization, hence, the target localization problem can be formulated using the basin of attraction of the local maxima. We employ a metric derived from the Bhattacharyya coefficient as similarity measure, and use the mean shift procedure to perform the optimization. In the presented tracking examples, the new method successfully coped with camera motion, partial occlusions, clutter, and target scale variations. Integration with motion filters and data association techniques is also discussed. We describe only a few of the potential applications: exploitation of background information, Kalman tracking using motion models, and face tracking.}, author = {Comaniciu, D. and Ramesh, V. and Meer, P. }, citeulike-article-id = {802362}, doi = {http://dx.doi.org/10.1109/TPAMI.2003.1195991}, journal = {Pattern Analysis and Machine Intelligence, IEEE Transactions on}, keywords = {tracking}, number = {5}, pages = {564--577}, posted-at = {2008-01-25 05:34:20}, priority = {2}, title = {Kernel-based object tracking}, url = {http://dx.doi.org/10.1109/TPAMI.2003.1195991}, volume = {25}, year = {2003} }

TY - JOUR ID - citeulike:802362 L3 - citeulike-article-id:802362 TI - Kernel-based object tracking JF - Pattern Analysis and Machine Intelligence, IEEE Transactions on VL - 25 IS - 5 SP - 564 EP - 577 N2 - A new approach toward target representation and localization, the central component in visual tracking of nonrigid objects, is proposed. The feature histogram-based target representations are regularized by spatial masking with an isotropic kernel. The masking induces spatially-smooth similarity functions suitable for gradient-based optimization, hence, the target localization problem can be formulated using the basin of attraction of the local maxima. We employ a metric derived from the Bhattacharyya coefficient as similarity measure, and use the mean shift procedure to perform the optimization. In the presented tracking examples, the new method successfully coped with camera motion, partial occlusions, clutter, and target scale variations. Integration with motion filters and data association techniques is also discussed. We describe only a few of the potential applications: exploitation of background information, Kalman tracking using motion models, and face tracking. KW - tracking AU - Comaniciu, D AU - Ramesh, V AU - Meer, P PY - 2003/// UR - http://dx.doi.org/10.1109/TPAMI.2003.1195991 ER -