ML Anthology

Time-Varying Shape Tensors for Scenes with Multiply Moving Points

Anat Levin, Lior Wolf, Amnon Shashua
CVPR 2001 pp. I:623-630
doi:10.1109/CVPR.2001.990532 /cvpr/2001/levin2001cvpr-time/

Abstract

We derive single view indexing functions for dynamic scenes - where dynamic is defined as a scene consisting of multiply moving points each moving independently with constant velocity. The indexing functions we derive are view independent and form a generalization of the "shape tensors" associated with rigid scenes by introducing a time-varying parameter We derive those indexing functions under full 3D projective, 3D affine, and various reduced configurations. The indexing functions were implemented and tested for matching against objects for which their non-rigid motion is an intrinsic part of their character - human gait recognition and hand gesture identification are the two chosen application examples.

CVPR Semantic Scholar

Cite

Text

Levin et al. "Time-Varying Shape Tensors for Scenes with Multiply Moving Points." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2001. doi:10.1109/CVPR.2001.990532

Markdown

[Levin et al. "Time-Varying Shape Tensors for Scenes with Multiply Moving Points." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2001.](https://mlanthology.org/cvpr/2001/levin2001cvpr-time/) doi:10.1109/CVPR.2001.990532

BibTeX

@inproceedings{levin2001cvpr-time,
  title     = {{Time-Varying Shape Tensors for Scenes with Multiply Moving Points}},
  author    = {Levin, Anat and Wolf, Lior and Shashua, Amnon},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2001},
  pages     = {I:623-630},
  doi       = {10.1109/CVPR.2001.990532},
  url       = {https://mlanthology.org/cvpr/2001/levin2001cvpr-time/}
}