RPN 2 DeepParts 22 part pool 2 3 HOG 12 LBP 13 Harr - like 15 DPM 16 Deformable Parts Model VGG16 X. Wang 14 VGG Convolutiona

Size: px

Start display at page:

Download "RPN 2 DeepParts 22 part pool 2 3 HOG 12 LBP 13 Harr - like 15 DPM 16 Deformable Parts Model VGG16 X. Wang 14 VGG Convolutiona"

甲卢
5 years ago
Views:

1 CHINESE JOURNAL OF STEREOLOGY AND IMAGE ANALYSIS Vol. 22 No. 2 June DOI /j RPN RPN RPN RPN TP37 A End - to - end monitoring scene pedestrian detection based on improved RPN YANG Dongming 1 2 XU Shibiao 3 MENG Weiliang 3 GE Shuiying 1 YANG Zhen 1 2 ZHANG Xiaopeng 3 1. National Science Library of CAS Beijing China 2. University of Chinese Academy of Sciences Beijing China 3. Institute of Automation of CAS Beijing China Abstract Compared to other scenes pedestrian detection in monitoring scenes gets larger flow of people and higher degree of occlusion. This paper presents an end - to - end detection scheme based on region proposal network RPN. On the one hand we improved RPN by combining with a deep convolution network designed by ourselves to obtain a new deep neural network for pedestrian detection. On the other hand we introduced head - shoulders model to improved detection performance for pedestrian detection at monitoring scenes which improves the detection speed. Eventually we achieved end - to - end detection. The experimental results show that the method effectively improved the detection performance by reducing the missing rate and speeding up the detection. Key words monitor pedestrian detection deep learning RPN network No. 2012BAH42F E - mail shuiying. ge@ rcsc. cas. cn

2 RPN 2 DeepParts 22 part pool 2 3 HOG 12 LBP 13 Harr - like 15 DPM 16 Deformable Parts Model VGG16 X. Wang 14 VGG Convolutional neural R - CNN Region of CNN 18 CNNs selective search region proposals SPP - net Fast - RCNN 20 SPP spatial pyramid pooling ROI region of interest pooling Girshick R Faster R - CNN 20 RPN Region Proposal Network RPN proposal ROI pooling Faster R - CNN RPN 5 RPN anchors R 3. 1 Benenson 7 S Zhang 8 state of art VGG16 23 Oxford Net Andrew Zisserman 2014 ILSVRC localization and classification networks CNNs 17 RPN CNNs Deep ConvNet RPN Classifier Deep JointDeep 9 ConvNet RPN R - CNN 1

$3 rpn_bbox_ 512 pred 2 RPN 4 t i t xi t yi t wi t hi K anchor anchor 4K anchor 2 b 3 3 300 512 3 3 L( { p }{ i t }) = 1 i N cls p i p * i + 1 i N reg p * i i L reg t i t * i 2 i i anchor p * i = 3 1$

3 RPN Deep ConvNet RPN Classifier 1 5 ground - truth IoU Intersection - over - 4 Max pooling union IoU ground - truth Relu IoU 0. 3 rpn_bbox_ 512 pred 2 RPN 4 t i t xi t yi t wi t hi K anchor anchor 4K anchor 2 b L( { p }{ i t }) = 1 i N cls p i p * i + 1 i N reg p * i i L reg t i t * i 2 i i anchor p * i = t * i ground - true box 1 1 anchor box a K anchor = RoI pooling RoI pooling N anchor = W Conv5 H Conv5 K anchor 1 ROIs region of interest W Conv5 H Conv5 rpn_cls_score rpn_bbox_pred rpn_cls_score resize P i 0 or 1 crop /wrap K anchor anchor 2K

212 2017 22 2 2 a b RPN 3 3 K anchor anchors 2 K anchor 4 K anchor 3.

Xeon VOC2007 25 INRIA Person dataset 26 ImageNet R CPU E5-2620 v3 @

4 a b RPN 3 3 K anchor anchors 2 K anchor 4 K anchor % 63% 98% 3 a Chen Q 23 SED - PD 3 a 3 b 4 3 a b CPU Intel R Xeon VOC INRIA Person dataset 26 ImageNet R CPU E GHz GPU 12G momentum 0. 9 weight decay TITAN X Pascal Caffe K train_loss /auc - iteatation RPN 4 train_loss - iteatation auc - iteatation s /iter INRIA Person dataset %

2017 22 2 RPN 213 0. 109 s 9. 17 Faster 1 INRIA Person dataset R - CNN VOC2007 + VOC2012 200 VOC2007 73. 2% 457 5 / 2 INRIA Person dataset 4 4.

5 RPN s Faster 1 INRIA Person dataset R - CNN VOC VOC VOC % / 2 INRIA Person dataset INRIA Person dataset a 3 a s INRIA Person dataset b 5 a 1 a conv5_2_relu5_2 rpn_cls_score rpn_bbox_pred rpn_output b

6 J DOI /j. cnki Zeng Min Zhou Yilong. Simulation of pedestrian detection based on deep learning model J. Journal of Nanjing Uni- of Posts and Telccommunications versity RPN DOI /j. cnki J DOI /j. issn Su Songzhi Li Shaozi Chen Shuyuan et al. A Survey on 2 pedestrian detection J. Acta Electronica Sinica 2012 VGG DOI /j. issn RPN Zeng Xingyu Ouyang Wanli Wang Xiaogang. Multi - stage contextual deep learning for pedestrian detection 5 7 Benenson R Omran M Hosang J et al. Ten years of pe- destrian detection what have we learned C / /ECCV RPN 2014 Workshops Zhang S Benenson R Omran M et al. How far are we from solving pedestrian detection C / / in CVPR Ouyang Wanli Wang Xiaogang. Joint deep learning for pedestrian detection C / / IEEE International Conference on Computer Vision Luo Ping Tian Yonglong Wang Xiaogang et al. Switchable deep network for pedestrian detection C / /IEEE Conference on Computer Vision and Pattern Recognition References 1. D Dalal N Triggs B. Histograms of oriented gradients for Wang Bin. Pedestrian detection based on Deep Learning human detection C / /IEEE Computer Society Conference D. Beijing Jiaotong University on Computer Vision & Pattern Recognition. IEEE Comput- 2. er Society D Liu Xiaohui. Pedestrian Detection and Tracking in Complex Visual Scenes D. Tianjin University Ojala T Pietikainen M Maenpaa T. Multiresolution gray - scale and rotation invariant texture classification with local binary patterns C / /PAMI HOG 14 Wang X Han T X Yan S. An HOG - LBP human detector D Zhou Kejia. Research on pedestrian detection and tracking with partial occlusion handling C / /IEEE International Conference on Computer Vision. IEEE based on HOG and template matching D. Jilin University C / / IEEE International Conference on Computer Vision J. Hosang M Omran R Benenson et al. Taking a deeper look at pedestrians C / /in CVPR Zhang S Bauckhage C Cremers A B. Informed Haar - Like Features Improve Pedestrian Detection C / /Comput-

7 RPN 215 er Vision and Pattern Recognition IEEE Felzenszwalb P Mcallester D Ramanan D. A discriminatively trained multiscale deformable part model C / / IEEE Computer Society Conference on Computer Vision and Pattern Recognition. DBLP Krizhevsky A Sutskever I Hinton G E. Imagenet classification with deep convolutional neural networks C / /Advances in Neural Information Processing Systems Girshick R Donahue J Darrell T et al. Rich feature hierarchies for accurate object detection and semantic segmentation C / /Computer Vision and Pattern Recognition CVPR 2014 IEEE Conference on. IEEE He K Zhang X Ren S et al. Spatial pyramid pooling in deep convolutional networics for visual recognition J. ArXiv Preprint ArXiv Girshick R. Fast R - CNN C / /IEEE International Conference on Computer Vision IEEE Computer Society Ren S He K Girshick R et al. Faster R - CNN Towards real - time object detection with region proposal networks J. IEEE Transactions on Pattern Analysis & Machine Intelligence Tian Y Luo P Wang X et al. Deep learning strong parts for pedestrian detection C / / IEEE International Conference on Computer Vision IEEE Simonyan K Zisserman A. Very deep convolutional networks for large - scale image recognition J. In ICLR Chen Q Jiang W Zhao Y et al. Part - based deep network for pedestrian detection in surveillance videos C / / IEEE Everingham M Gool L Williams C K et al. The Pascal visual object classes VOC challenge J. International Journal of Computer Vision Dalal N Triggs B. Histograms of oriented gradients for human detection C / /CVRR 2015 France

第 2 期王向东等 : 一种运动轨迹引导下的举重视频关键姿态提取方法 257 竞技体育比赛越来越激烈, 为了提高体育训练的效率, 有必要在体育训练中引入科学定量的方法许多

第 2 期王向东等 : 一种运动轨迹引导下的举重视频关键姿态提取方法 257 竞技体育比赛越来越激烈, 为了提高体育训练的效率, 有必要在体育训练中引入科学定量的方法许多 2014 年 4 月图学学报 April 2014 第 35 卷第 2 期 JOURNAL OF GRAPHICS Vol.35 No.2 一种运动轨迹引导下的举重视频关键姿态提取方法王向东 1, 张静文 2, 毋立芳 2, 徐文泉 (1. 国家体育总局体育科学研究所, 北京 100061;2. 北京