Action Recognition based on Subdivision-Fusion Model

Zongbo Hao, Linlin Lu, Qianni Zhang, Jie Wu, Ebroul Izquierdo, Juanyu Yang and Jun Zhao

Abstract

This paper proposes a novel Subdivision-Fusion Model (SFM) to recognize human actions. In most action recognition tasks, overlapping feature distribution is a common problem leading to overfitting. In the subdivision stage of the proposed SFM, samples in each category are clustered. Then, such samples are grouped into multiple more concentrated subcategories. Boundaries for the subcategories are easier to find and as consequence overfitting is avoided. In the subsequent fusion stage, the multi-subcategories classification results are converted back to the original category recognition problem. Two methods to determine the number of clusters are provided. The proposed model has been thoroughly tested with four popular datasets. In the Hollywood2 dataset, an accuracy of 79.4% is achieved, outperforming the state-of-the-art accuracy of 64.3%. The performance on the YouTube Action dataset has been improved from 75.8% to 82.5%, while considerably improvements are also observed on the KTH and UCF50 datasets.

Session

Poster 1

Files

Extended Abstract (PDF, 445K)
Paper (PDF, 934K)

DOI

10.5244/C.29.50
https://dx.doi.org/10.5244/C.29.50

Citation

Bibtex

@inproceedings{BMVC2015_50,
	title={Action Recognition based on Subdivision-Fusion Model},
	author={Zongbo Hao and Linlin Lu and Qianni Zhang and Jie Wu and Ebroul Izquierdo and Juanyu Yang and Jun Zhao},
	year={2015},
	month={September},
	pages={50.1-50.12},
	articleno={50},
	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.29.50},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.50}
}