DocumentCode :
254084
Title :
The Fastest Deformable Part Model for Object Detection
Author :
Junjie Yan ; Zhen Lei ; Longyin Wen ; Li, Stan Z.
Author_Institution :
Nat. Lab. of Pattern Recognition, Inst. of Autom., Beijing, China
fYear :
2014
fDate :
23-28 June 2014
Firstpage :
2497
Lastpage :
2504
Abstract :
This paper solves the speed bottleneck of deformable part model (DPM), while maintaining the accuracy in detection on challenging datasets. Three prohibitive steps in cascade version of DPM are accelerated, including 2D correlation between root filter and feature map, cascade part pruning and HOG feature extraction. For 2D correlation, the root filter is constrained to be low rank, so that 2D correlation can be calculated by more efficient linear combination of 1D correlations. A proximal gradient algorithm is adopted to progressively learn the low rank filter in a discriminative manner. For cascade part pruning, neighborhood aware cascade is proposed to capture the dependence in neighborhood regions for aggressive pruning. Instead of explicit computation of part scores, hypotheses can be pruned by scores of neighborhoods under the first order approximation. For HOG feature extraction, look-up tables are constructed to replace expensive calculations of orientation partition and magnitude with simpler matrix index operations. Extensive experiments show that (a) the proposed method is 4 times faster than the current fastest DPM method with similar accuracy on Pascal VOC, (b) the proposed method achieves state-of-the-art accuracy on pedestrian and face detection task with frame-rate speed.
Keywords :
approximation theory; correlation methods; feature extraction; filtering theory; gradient methods; matrix algebra; object detection; 1D correlations; 2D correlation; DPM; HOG feature extraction; Pascal VOC; aggressive pruning; cascade part pruning; cascade version; deformable part model; face detection task; feature map; first order approximation; frame-rate speed; hypothesis pruning; linear combination; look-up tables; matrix index operations; neighborhood aware cascade; object detection; orientation partition; pedestrian detection task; proximal gradient algorithm; rank filter; root filter; Acceleration; Accuracy; Correlation; Deformable models; Feature extraction; Table lookup; Training; deformable part model; face detection; object detection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on
Conference_Location :
Columbus, OH
Type :
conf
DOI :
10.1109/CVPR.2014.320
Filename :
6909716
Link To Document :
بازگشت