Learning to Track: Conceptual Manifold mAP for Closed-Form Tracking

Elgammal, Ahmed M.

doi:10.1109/CVPR.2005.209

Learning to Track: Conceptual Manifold mAP for Closed-Form Tracking

Ahmed M. Elgammal

CVPR 2005 pp. 724-730

doi:10.1109/CVPR.2005.209 /cvpr/2005/elgammal2005cvpr-learning/

Abstract

Our objective is to model the visual manifold of object appearance corresponding to geometric transformation. We learn a generative model for object appearance where the appearance of the object at each new frame is a function that maps from a conceptual representation of the geometric transformation space into the visual manifold. By learning such generative model we can infer the geometric transformation (track) directly from the tracked object appearance. As a result tracking can be achieved in a closed form and therefore can be done very efficiently.

PDF CVPR Semantic Scholar

Cite

Text

Elgammal. "Learning to Track: Conceptual Manifold mAP for Closed-Form Tracking." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2005. doi:10.1109/CVPR.2005.209

Markdown

[Elgammal. "Learning to Track: Conceptual Manifold mAP for Closed-Form Tracking." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2005.](https://mlanthology.org/cvpr/2005/elgammal2005cvpr-learning/) doi:10.1109/CVPR.2005.209

BibTeX

@inproceedings{elgammal2005cvpr-learning,
  title     = {{Learning to Track: Conceptual Manifold mAP for Closed-Form Tracking}},
  author    = {Elgammal, Ahmed M.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2005},
  pages     = {724-730},
  doi       = {10.1109/CVPR.2005.209},
  url       = {https://mlanthology.org/cvpr/2005/elgammal2005cvpr-learning/}
}