Gradient Boundary Histograms for Action Recognition

Feng Shi, Robert Laganière, Emil M. Petriu

WACV 2015 pp. 1107-1114

doi:10.1109/WACV.2015.152 /wacv/2015/shi2015wacv-gradient/

Abstract

This paper introduces a high efficient local spatiotemporal descriptor, called gradient boundary histograms (GBH). The proposed GBH descriptor is built on simple spatio-temporal gradients, which are fast to compute. We demonstrate that it can better represent local structure and motion than other gradient-based descriptors, and significantly outperforms them on large realistic datasets. A comprehensive evaluation shows that the recognition accuracy is preserved while the spatial resolution is greatly reduced, which yields both high efficiency and low memory usage.

PDF WACV Semantic Scholar

Cite

Text

Shi et al. "Gradient Boundary Histograms for Action Recognition." IEEE/CVF Winter Conference on Applications of Computer Vision, 2015. doi:10.1109/WACV.2015.152

Markdown

[Shi et al. "Gradient Boundary Histograms for Action Recognition." IEEE/CVF Winter Conference on Applications of Computer Vision, 2015.](https://mlanthology.org/wacv/2015/shi2015wacv-gradient/) doi:10.1109/WACV.2015.152

BibTeX

@inproceedings{shi2015wacv-gradient,
  title     = {{Gradient Boundary Histograms for Action Recognition}},
  author    = {Shi, Feng and Laganière, Robert and Petriu, Emil M.},
  booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  year      = {2015},
  pages     = {1107-1114},
  doi       = {10.1109/WACV.2015.152},
  url       = {https://mlanthology.org/wacv/2015/shi2015wacv-gradient/}
}