Real-Time Pedestrian Detection with Deep Network Cascades

Anelia Angelova, Alex Krizhevsky, Vincent Vanhoucke, Abhijit Ogale and Dave Ferguson

Abstract

We present a new real-time approach to object detection that exploits the efficiency of cascade classifiers with the accuracy of deep neural networks. Deep networks have been shown to excel at classification tasks, and their ability to operate on raw pixel input without the need to design special features is very appealing. However, deep nets are notoriously slow at inference time. In this paper, we propose an approach that cascades deep nets and fast features, that is both very fast and very accurate. We apply it to the challenging task of pedestrian detection. Our algorithm runs in real-time at 15 frames per second. The resulting approach achieves a 26.2 percent average miss rate on the Caltech Pedestrian detection benchmark, which is competitive with the very best reported results. It is the first work we are aware of that achieves very high accuracy while running in real-time.

Session

Poster 1

Files

Extended Abstract (PDF, 115K)
Paper (PDF, 2M)

DOI

10.5244/C.29.32
https://dx.doi.org/10.5244/C.29.32

Citation

Bibtex

@inproceedings{BMVC2015_32,
	title={Real-Time Pedestrian Detection with Deep Network Cascades},
	author={Anelia Angelova and Alex Krizhevsky and Vincent Vanhoucke and Abhijit Ogale and Dave Ferguson},
	year={2015},
	month={September},
	pages={32.1-32.12},
	articleno={32},
	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.29.32},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.32}
}