APT: Action localization proposals from dense trajectories

Jan C. van Gemert, Mihir Jain, Ella Gati and Cees G. M. Snoek

Abstract

This paper is on action localization in video with the aid of spatio-temporal proposals. To alleviate the computational expensive video segmentation step of existing proposals, we propose bypassing the segmentations completely by generating proposals directly from the dense trajectories used to represent videos during classification. Our Action localization Proposals from dense Trajectories (APT) uses an efficient proposal generation algorithm to handle the high number of trajectories in a video. Our spatio-temporal proposals are faster than current methods and outperform the localization and classification accuracy of current proposals on UCF Sports, UCF 101, and MSR-II video datasets.

Session

Action and Event

Files

PDF iconExtended Abstract (PDF, 2004K)
PDF iconPaper (PDF, 2M)

DOI

10.5244/C.29.177
https://dx.doi.org/10.5244/C.29.177

Citation

Jan C. van Gemert, Mihir Jain, Ella Gati and Cees G. M. Snoek. APT: Action localization proposals from dense trajectories. In Xianghua Xie, Mark W. Jones, and Gary K. L. Tam, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 177.1-177.12. BMVA Press, September 2015.

Bibtex

@inproceedings{BMVC2015_177,
	title={APT: Action localization proposals from dense trajectories},
	author={Jan C. van Gemert and Mihir Jain and Ella Gati and Cees G. M. Snoek},
	year={2015},
	month={September},
	pages={177.1-177.12},
	articleno={177},
	numpages={12},
	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
	publisher={BMVA Press},
	editor={Xianghua Xie, Mark W. Jones, and Gary K. L. Tam},
	doi={10.5244/C.29.177},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.177}
}