6 : 835, RGBD.,.,,,,., (Mean shift) OpenCV.,.,,.. 1. Fig. 1 1 The framework of video object tracking.,,,.,.,,,,. 1.1,.,, : 1) (Mean shift),. [1],,, Bh

Size: px

Start display at page:

Download "6 : 835, RGBD.,.,,,,., (Mean shift) OpenCV.,.,,.. 1. Fig. 1 1 The framework of video object tracking.,,,.,.,,,,. 1.1,.,, : 1) (Mean shift),. [1],,, Bh"

鼐历
7 years ago
Views:

1 42 6 Vol. 42, No ACTA AUTOMATICA SINICA June, ,...,.. DOI,,,,,,.., 2016, 42(6): /j.aas.2016.c Advances on Application of Deep Learning for Video Object Tracking GUAN Hao 1 XUE Xiang-Yang 1 AN Zhi-Yong 1 Abstract Video object tracking is an important research topic of computer vision with numerous applications including surveillance, robotics, human-computer interface, etc. The coming of big data era and the rise of deep learning methods have offered new opportunities for the research of tracking. Firstly, we present the general framework for video object tracking research. Then, we introduce new arisen datasets and evaluation methodology. We highlight the application of the rapid-developing deep-learning methods including stacked autoencoder and convolutional neural network on video object tracking. Finally, we have a discussion and provide insights for future. Key words Object tracking, video analysis, online learning, deep learning, big data Citation Guan Hao, Xue Xiang-Yang, An Zhi-Yong. Advances on application of deep learning for video object tracking. Acta Automatica Sinica, 2016, 42(6): ,, ( ). : ( ),,. ( ),.,, Manuscript received October 26, 2015; accepted May 3, 2016 ( ), ( ) Supported by National Natural Science Foundation of China ( ) and Science and Technology Commission of Shanghai Municipality ( ) Recommended by Associate Editor KE Deng-Feng Shanghai Key Laboratory of Intelligent Information Processing, School of Computer Science, Fudan University, Shanghai ( ).,.,..,,,,.,., ( ).,.,,.,,

2 6 : 835, RGBD.,.,,,,., (Mean shift) OpenCV.,.,,.. 1. Fig. 1 1 The framework of video object tracking.,,,.,.,,,,. 1.1,.,, : 1) (Mean shift),. [1],,, Bhattacharyya,. 2) (Slide window),.,. 3) (Particle filter) [2], ( ).,., 1.2,.,,

3 : 1) ; 2),..,,.,, :, (Haar ).,.,.. Haar, [3]. Haar,. : [4 5] ; Color name [6].,,, (Local binary pattern, LBP) [7].,,.. [8 9] SIFT (Scale invariant feature transform) SURF (Speeded up robust features),. HOG (Histogram of oriented gradient), [10]. HOG,.,..,, (Optical flow),,. LK [11] HS [12].,.,.,, TLD (Tracking learning detection) [13 14] ,. (Principle component analysis, PCA), (Deep learning),,.,,., 1.3 [15 16].,,. : [17].,

4 6 : 837,. : [18 20] [21 22] [23 28]., [27],. [28] L 1. Fig. 2 2 The generative appearance model,, ( ) (Tracking by detection),..,,, ( ), 3. Fig. 3 3 The discriminative appearance model,,,,.,, [29 31] Boosting [32 34] [35] [36 37] [13] [38]. 1.4 ( ),,,, ;,,., : 1).,., 2) 3) ( )..,,, 1). 4),., 2 2.1, [39], ImageNet [40] Pascal VOC [41] TRECVID [42]. [43] VTB,. 50, 100 [44].,,. VOT [45]. Pascal VOC, 2013 VTB,.

5 :. (Accuracy):.,. : 1) (Deviation): ; 2) (False positive): ; 3) (False negative): (Robustness):,.,. (Efficiency):.,.. 1) (Center location error):.,. 2) (Overlap rate): S T, S G,, : R = area(s T S G ) area(s T S G,,. ) 3) (Success rate):,. 4) (Precision plot) (Success plot): 3),, ;. 5) (Temporal robustness evaluation, TRE) (Spatial robustness evaluation, SRE). [43] TRE,,. SRE,,, 6) FPS (Frames per second):, (Deep learning),,,, 20 80, Rumelhart Hinton Williams (Back propagation, BP) [46].. BP, (Support vector machine, SVM) Hinton [47]., [48], [49],. [50 51], [52 53].,.,, ( )., Fig. 4 4 The framework of recognition in multimedia, HOG LBP,,.,

6 6 : 839 [54 55],.,,,, ; GPU, GPU, Caffe [56] Theano [57] GPU. 3.2 (Deep belief net, DBN) [58] (Stacked autoencoder) [59]. (Multilayer perceptron) (Convolutional neural network, CNN) [60]., (Feedforward neural network, FNN) (Recurrent neural network, RNN) [61],.,. (Long and short term memory, LSTM) [62 64],,,, Fig. 5 5 The basic models of deep learning 3.3.,, : 1),,,, 2),,,., : 1),,,, 2),, , (Autoencoder).,,,. (Stacked autoencoder),,,.

7 (Logistic regression). Fig. 6 6 The illustration of autoencoder [59]...,, Fig. 7 7 The illustration of denoise autoencoder 3.4.2, [65]. [66]...,, 1, 0,.,,., ( ), 8 [65] Fig. 8 Denoise autoencoder for video tracking [65],,..., +.,. : 1),,.. 2),.,,. 3),, [67] [65], AdaBoost, 4.,, [68], [65], [68].

8 6 : 841, (Independent subspace analysis, ISA) (Temporal slowness constraint).,. [65]. [69]., [70]. : 1),,,. 2),,,,.,,,, , (Convolutional neural network, CNN)., Hubel Wiesel. (Receptive field) [71]., [72] (Neocognitron), LeCun BP [60, 73], [74]., ImageNet,, [49], Krizhevsky,. ( C1, C2 ),, (Pooling, ),, Fig. 9 [60, 73] 9 The illustration of convolutional neural [60, 73] network.,, [75 76], 3.5.2, :, ;, [77].. [78].,..,. [79] 7

9 (1, 0 ),.. ImageNet,. [80],,, [81 82],.,, [83 84],.,,,., [85],, ,. Haar., [86]., HOG,,,,,,,. [87], [29] Struck, ,,,,..,. 10 ( [81])., 10 (a), 10 (b),, ( 2.2 ).,.,, [65]. 3.7,,,,

10 6 : 843. : 10 [81] (FCNT, DLT [65].) Fig. 10 Comparison of CNN-based tracking method and other trackers [81] (FCNT is a CNN-based tracker and DLT is an autoencoder-based tracker [65].) 1),,.,. ImageNet,.,, 2),..,. 3), 0 1,,,.,. 4).,,.,, 5),,,,,,, [88]. 4,,,,.,,

11 : 1),, 2),.,,,. 3), 4),,,. 5,.,.,. References 1 Comaniciu D, Ramesh V, Meer P. Real-time tracking of nonrigid objects using mean shift. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition. Hilton Head Island, SC: IEEE, Risfic B, Arulampalam S, Gordon N. Beyond the Kalman filter-book review. IEEE Aerospace and Electronic Systems Magazine, 2004, 19(7): Viola P, Jones M. Rapid object detection using a boosted cascade of simple features. In: Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Hawaii, USA: IEEE, I-511 I Pérez P, Hue C, Vermaak J, Gangnet M. Color-based probabilistic tracking. In: Proceedings of the 7th European Conference on Computer Vision. Copenhagen, Denmark: Springer, Possegger H, Mauthner T, Bischof H. In defense of colorbased model-free tracking. In: Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE, Danelljan M, Khan F S, Felsberg M, van de Weijer J. Adaptive color attributes for real-time visual tracking. In: Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA: IEEE, Ojala T, Pietikainen M, Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions. In: Proceedings of the 12th IAPR International Conference on Pattern Processing. Jerusalem: IEEE, Zhou H Y, Yuan Y, Shi C M. Object tracking using SIFT features and mean shift. Computer Vision and Image Understanding, 2009, 113(3): Miao Q, Wang G J, Shi C B, Lin X G, Ruan Z W. A new framework for on-line object tracking based on SURF. Pattern Recognition, 2011, 32(13): Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA: IEEE, Lucas B D, Kanade T. An iterative image registration technique with an application to stereo vision. In: Proceedings of the 7th International Joint Conference on Artificial Intelligence. San Francisco, CA, USA: Morgan Kaufmann Publishers Inc., Horn B K P, Schunck B G. Determining optical flow. Artificial Intelligence, 1981, 17(2): Kalal Z, Mikolajczyk K, Matas J. Tracking-learningdetection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7): Kalal Z, Mikolajczyk K, Matas J. Forward-backward error: automatic detection of tracking failures. In: Proceedings of the 20th IEEE International Conference on Pattern Recognition. Istanbul: IEEE, Li X, Hu W M, Shen C H, Zhang Z F, Dick A, van den Hengel A. A survey of appearance models in visual object tracking. ACM Transactions on Intelligent Systems and Technology, 2013, 4(4): Article No Zhang Huan-Long, Hu Shi-Qiang, Yang Guo-Sheng. Video object tracking based on appearance models learning. Journal of Computer Research and Development, 2015, 52(1): (,,.., 2015, 52(1): ) 17 Hou Zhi-Qiang, Han Chong-Zhao. A survey of visual tracking. Acta Automatica Sinica, 2006, 32(4): (,.., 2006, 32(4): )

12 6 : Adam A, Rivlin E, Shimshoni I. Robust fragments-based tracking using the integral histogram. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, NY, USA: IEEE, Alt N, Hinterstoisser S, Navab N. Rapid selection of reliable templates for visual tracking. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, CA, USA: IEEE, He S F, Yang Q X, Lau R W H, Wang J, Yang M H. Visual tracking via locality sensitive histograms. In: Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland, OR, USA: IEEE, Black M J, Jepson A D. EigenTracking: robust matching and tracking of articulated objects using a view-based representation. International Journal of Computer Vision, 1998, 26(1): Ross D A, Lim J, Lin R S, Yang M H. Incremental learning for robust visual tracking. International Journal of Computer Vision, 2008, 77(1 3): Zhang T Z, Liu S, Xu C S, Yan S C, Ghanem B, Ahuja N, Yang M H. Structural sparse tracking. In: Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE, Jia X, Lu H C, Yang M H. Visual tracking via adaptive structural local sparse appearance model. In: Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA: IEEE, Zhang T Z, Ghanem B, Liu S, Ahuja N. Robust visual tracking via multi-task sparse learning. In: Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA: IEEE, Zhang S P, Yao H X, Sun X, Lu X S. Sparse coding based visual tracking: review and experimental comparison. Pattern Recognition, 2013, 46(7): Wright J, Ma Y, Mairal J, Sapiro G, Huang T S, Yan S C. Sparse representation for computer vision and pattern recognition. Proceedings of the IEEE, 2010, 98(6): Mei X, Ling H B. Robust visual tracking using L 1 minimization. In: Proceedings of the 12th IEEE International Conference on Computer Vision. Kyoto: IEEE, Hare S, Saffari A, Torr P H S. Struck: structured output tracking with kernels. In: Proceedings of the 2011 IEEE International Conference on Computer Vision. Barcelona: IEEE, Avidan S. Support vector tracking. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(8): Bai Y C, Tang M. Robust tracking via weakly supervised ranking SVM. In: Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA: IEEE, Grabner H, Grabner M, Bischof H. Real-time tracking via on-line boosting. In: Proceedings of the British Machine Vision Conference. Edinburgh, UK: BMVA Press, Grabner H, Leistner C, Bischof H. Semi-supervised on-line boosting for robust tracking. In: Proceedings of the 10th European Conference on Computer Vision. Marseille, France: Springer, Stalder S, Grabner H, van Gool L. Beyond semi-supervised tracking: tracking should be as simple as detection, but not simpler than recognition. In: Proceedings of the 12th IEEE International Conference on Computer Vision Workshops. Kyoto: IEEE, Babenko B, Yang M H, Belongie S. Visual tracking with online multiple instance learning. In: Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, USA: IEEE, Henriques J F, Caseiro R, Martins P, Batista J. Exploiting the circulant structure of tracking-by-detection with kernels. In: Proceedings of the 12th European Conference on Computer Vision. Florence, Italy: Springer, Henriques J F, Caseiro R, Martins P, Batista J. High-speed tracking with kernelized correlation filters. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(3): Zhang K H, Zhang L, Yang M H. Real-time compressive tracking. In: Proceedings of 12th European Conference on Computer Vision. Florence, Italy: Springer, Huang Kai-Qi, Ren Wei-Qiang, Tan Tie-Niu. A review on image object classification and detection. Chinese Journal of Computers, 2014, 37(6): (,,., 2014, 37(6): ) 40 Deng J, Dong W, Socher R, Li J J, Li K, Li F F. ImageNet: a large-scale hierarchical image database. In: Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, USA: IEEE, Everingham M, Van Gool L, Williams C K I, Winn J, Zisserman A. The PASCAL visual object classes (VOC) challenge. International Journal of Computer Vision, 2010, 88(2): Smeaton A F, Over P, Kraaij W. Evaluation campaigns and TRECVid. In: Proceedings of the 8th ACM International Workshop on Multimedia Information Retrieval. Santa Barbara, CA, USA: ACM, Wu Y, Lim J, Yang M H. Online object tracking: a benchmark. In: Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland, OR, USA: IEEE, Wu Y, Lim J, Yang M H. Object tracking benchmark. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): Kristan M, Matas J, Leonardis A, Felsberg M, Cehovin L, Fernández G, Vojír T, Häger G, Nebehay G, Pflugfelder R. The visual object tracking VOT2015 challenge results. In: Proceedings of the 2015 IEEE International Conference on Computer Vision Workshops. Santiago: IEEE, Rumelhart D E, Hinton G E, Williams R J. Learning representations by back-propagating errors. Nature, 1986, 323(6088): Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks. Science, 2006, 313(5786):

13 Hinton G, Deng L, Yu D, Dahl G E, Mohamed A R, Jaitly N, Senior A, Vanhoucke V, Nguyen P, Sainath T N, Kingsbury B. Deep neural networks for acoustic modeling in speech recognition: the shared views of four research groups. IEEE Signal Processing Magazine, 2012, 29(6): Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks. In: Proceeding of Advances in Neural Information Processing Systems. Nevada, USA: MIT Press, Girshick R, Donahue J, Darrell T, Malik J. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA: IEEE, Ren S Q, He K M, Girshick R, Sun J. Faster R-CNN: towards real-time object detection with region proposal networks. In: Proceeding of Advances in Neural Information Processing Systems. Montréal, Canada: MIT Press, Karpathy A, Toderici G, Shetty S, Leung T, Sukthankar R, Li F F. Large-scale video classification with convolutional neural networks. In: Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA: IEEE, Ji S W, Xu W, Yang M, Yu K. 3D convolutional neural networks for human action recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(1): Lee T S, Mumford D, Romero R, Lamme V A F. The role of the primary visual cortex in higher level vision. Vision Research, 1998, 38(15 16): Lee T S, Mumford D. Hierarchical Bayesian inference in the visual cortex. Journal of the Optical Society of America A: Optics Image Science and Vision, 2003, 20(7): Jia Y Q, Shelhamer E, Donahue J, Karayev S, Long J, Girshick R, Guadarrama S, Darrell T. Caffe: convolutional architecture for fast feature embedding. In: Proceedings of the 22nd ACM International Conference on Multimedia. Orlando, FL, USA: ACM, Bergstra J, Bastien F, Breuleux O, Lamblin P, Pascanu R, Delalleau O, Desjardins G, Warde-Farley D, Goodfellow I J, Bergeron A, Bengio Y. Theano: deep learning on GPUS with python. In: Advances in Neural Information Processing Systems Workshops. Granada, Spain: MIT Press, Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets. Neural Computation, 2006, 18(7): Vincent P, Larochelle H, Bengio Y, Manzagol P A. Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th International Conference on Machine Learning. Helsinki, Finland: ACM, LeCun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 1998, 86(11): Jozefowicz R, Zaremba W, Sutskever I. An empirical exploration of recurrent network architectures. In: Proceedings of the 32nd International Conference on Machine Learning. Lille, France: JMLR, Hochreiter S, Schmidhuber J. Long short-term memory. Neural Computation, 1997, 9(8): Gers F A, Schraudolph N N, Schmidhuber J. Learning precise timing with LSTM recurrent networks. The Journal of Machine Learning Research, 2003, 3: Graves A, Liwicki M, Fernández S, Bertolami R, Bunke H, Schmidhuber J. A novel connectionist system for unconstrained handwriting recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(5): Wang N Y, Yeung D Y. Learning a deep compact image representation for visual tracking. In: Proceeding of Advances in Neural Information Processing Systems. Nevada, USA: MIT Press, Torralba A, Fergus R, Freeman W T. 80 million tiny images: a large data set for nonparametric object and scene recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(11): Zhou X Z, Xie L, Zhang P, Zhang Y N. An ensemble of deep neural networks for object tracking. In: Proceedings of the 2014 IEEE International Conference on Image Processing. Paris, France: IEEE, Kuen J, Lim K M, Lee C P. Self-taught learning of a deep invariant representation for visual tracking via temporal slowness principle. Pattern Recognition, 2015, 48(10): Ding J W, Huang Y Z, Liu W, Huang K Q. Severely blurred object tracking by learning deep image representations. IEEE Transactions on Circuits and Systems for Video Technology, 2016, 26(2): Dai L, Zhu Y S, Luo G B, He C. A low-complexity visual tracking approach with single hidden layer neural networks. In: Proceedings of the 13th IEEE International Conference on Control Automation Robotics and Vision. Singapore: IEEE, Hubel D H, Wiesel T N. Receptive fields, binocular interaction and functional architecture in the cat s visual cortex. Journal of Physiology, 1962, 160(1): Fukushima K. Neocognitron: a self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 1980, 36(4): LeCun Y, Boser B, Denker J S, Henderson D, Howard R E, Hubbard W, Jackel L D. Backpropagation applied to handwritten zip code recognition. Neural Computation, 1989, 1(4): LeCun Y, Kavukcuoglu K, Farabet C. Convolutional networks and applications in vision. In: Proceedings of 2010 IEEE International Symposium on Circuits and Systems. Paris, France: IEEE, Szegedy C, Liu W, Jia Y Q, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A. Going deeper with convolutions. In: Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE, Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arxiv: , 2014.

14 6 : Jin J, Dundar A, Bates J, Farabet C, Culurciello E. Tracking with deep neural networks. In: Proceedings of the 47th Annual Conference on Information Sciences and Systems (CISS). Baltimore, MD, USA: IEEE, Wang L, Liu T, Wang G, Chan K L, Yang Q X. Video tracking using learned hierarchical features. IEEE Transactions on Image Processing, 2015, 24(4): Wang N Y, Li S Y, Gupta A, Yeung D Y. Transferring rich feature hierarchies for robust visual tracking. arxiv: , Hong S, You T, Kwak S, Han B. Online tracking by learning discriminative saliency map with convolutional neural network. In: Proceedings of the 32th International Conference on Machine Learning. Lille, France: JMLR, Wang L J, Ouyang W L, Wang X G, Lu H C. Visual tracking with fully convolutional networks. In: Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago: IEEE, Ma C, Huang J B, Yang X K, Yang M H. Hierarchical convolutional features for visual tracking. In: Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago: IEEE, Li H X, Li Y, Porikli F. DeepTrack: learning discriminative feature representations online for robust visual tracking. IEEE Transactions on Image Processing, 2016, 25(4): Li H X, Li Y, Porikli F. Robust online visual tracking with a single convolutional neural network. In: Proceedings of the 12th Asian Conference on Computer Vision. Singapore: Springer, He Y, Dong Z, Yang M, Chen L, Pei M T, Jia Y D. Visual tracking using multi-stage random simple features. In: Proceedings of the 22nd International Conference on Pattern Recognition. Stockholm: IEEE, Danelljan M, Häger G, Khan F S, Felsberg M. Convolutional features for correlation filter based visual tracking. In: Proceedings of the 2015 IEEE International Conference on Computer Vision Workshop. Santiago: IEEE, Wang N Y, Shi J P, Yeung D Y, Jia J Y. Understanding and diagnosing visual tracking systems. In: Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago: IEEE, Hong Z B, Chen Z, Wang C H, Mei X, Prokhorov D, Tao D C. MUlti-Store tracker (MUSTer): a cognitive psychology inspired approach to object tracking. In: Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA: IEEE, ,.. guanh13@fudan.edu.cn (GUAN Hao Ph. D. candidate at the School of Computer Science, Fudan University. His research interest covers video analysis and deep learning. Corresponding author of this paper.).,,. xyxue@fudan.edu.cn (XUE Xiang-Yang Professor at the School of Computer Science, Fudan University. His research interest covers big video data analysis, computer vision, and deep learning.) azytyut@163.com (AN Zhi-Yong Postdoctor at the School of Computer Science, Fudan University. He received his Ph. D. degree from Xidian University in His research interest covers image and video content analysis and retrieval.)

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

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 2017 22 2 CHINESE JOURNAL OF STEREOLOGY AND IMAGE ANALYSIS Vol. 22 No. 2 June 2017 209 1007-1482 2017 02-0209 -0215 DOI 10. 13505 /j. 1007-1482. 2017. 22. 02. 014 RPN 1 2 3 3 1 1 2 3 1. 100090 2. 100090