Beyond Semi-Supervised Tracking: Tracking Should Be as Simple as Detection, but Not Simpler than Recognition
Abstract
We present a multiple classifier system for model-free tracking. The tasks of detection (finding the object of interest), recognition (distinguishing similar objects in a scene), and tracking (retrieving the object to be tracked) are split into separate classifiers in the spirit of simplifying each classification task. The supervised and semi-supervised classifiers are carefully trained on-line in order to increase adaptivity while limiting accumulation of errors, i.e. drifting. In the experiments, we demonstrate real-time tracking on several challenging sequences, including multi-object tracking of faces, humans, and other objects. We outperform other on-line tracking methods especially in case of occlusions and presence of similar objects. ©2009 IEEE.
Cite
Text
Stalder et al. "Beyond Semi-Supervised Tracking: Tracking Should Be as Simple as Detection, but Not Simpler than Recognition." IEEE/CVF International Conference on Computer Vision Workshops, 2009. doi:10.1109/ICCVW.2009.5457445Markdown
[Stalder et al. "Beyond Semi-Supervised Tracking: Tracking Should Be as Simple as Detection, but Not Simpler than Recognition." IEEE/CVF International Conference on Computer Vision Workshops, 2009.](https://mlanthology.org/iccvw/2009/stalder2009iccvw-beyond/) doi:10.1109/ICCVW.2009.5457445BibTeX
@inproceedings{stalder2009iccvw-beyond,
title = {{Beyond Semi-Supervised Tracking: Tracking Should Be as Simple as Detection, but Not Simpler than Recognition}},
author = {Stalder, Severin and Grabner, Helmut and Van Gool, Luc},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2009},
pages = {1409-1416},
doi = {10.1109/ICCVW.2009.5457445},
url = {https://mlanthology.org/iccvw/2009/stalder2009iccvw-beyond/}
}