›› 2015,Vol. 30 ›› Issue (2): 364-372.doi: 10.1007/s11390-015-1529-y

所属专题: Artificial Intelligence and Pattern Recognition Computer Graphics and Multimedia

• Special Section on Selected Paper from NPC 2011 • 上一篇    下一篇

基于断裂与连接的原始轨迹优化方法

Chun-Chao Guo(郭春超), Student Member, CCF, IEEE, Xiao-Jun Hu(胡晓军) Jian-Huang Lai*(赖剑煌), Member, CCF, IEEE, Shi-Chang Shi(石世昌), Shi-Zhe Chen(陈世哲)   

  1. School of Information Science and Technology, Sun Yat-sen University, Guangzhou 510006, China
  • 收稿日期:2014-12-26 修回日期:2015-01-13 出版日期:2015-03-05 发布日期:2015-03-05
  • 作者简介:Chun-Chao Guo received his B.E. degree in communication engineering with honors from Lanzhou University, Lanzhou, in 2010. He is currently working toward his Ph.D. degree in computer science at Sun Yat-sen University, Guangzhou. His research interests are in computer vision and pattern recognition, with a focus on human identity recognition, object tracking, object detection, and visual surveillance. He is a recipient of the Excellent Paper Award at the 2014 National Conference on Image and Graphics. He won the first prize in the 2014 National Graduate Contest on Smart-City Technology, and is one of the winners in the 2014 Bocom Cup Contest on Video Analysis. He is a student member of CCF and IEEE.
  • 基金资助:

    This work was supported by the National Natural Science Foundation of China under Grant No. 61173084 and the National Science & Technology Pillar Program of China under Grant No. 2012BAK16B06.

Raw Trajectory Rectification via Scene-Free Splitting and Stitching

Chun-Chao Guo(郭春超), Student Member, CCF, IEEE, Xiao-Jun Hu(胡晓军) Jian-Huang Lai*(赖剑煌), Member, CCF, IEEE, Shi-Chang Shi(石世昌), Shi-Zhe Chen(陈世哲)   

  1. School of Information Science and Technology, Sun Yat-sen University, Guangzhou 510006, China
  • Received:2014-12-26 Revised:2015-01-13 Online:2015-03-05 Published:2015-03-05
  • About author:Chun-Chao Guo received his B.E. degree in communication engineering with honors from Lanzhou University, Lanzhou, in 2010. He is currently working toward his Ph.D. degree in computer science at Sun Yat-sen University, Guangzhou. His research interests are in computer vision and pattern recognition, with a focus on human identity recognition, object tracking, object detection, and visual surveillance. He is a recipient of the Excellent Paper Award at the 2014 National Conference on Image and Graphics. He won the first prize in the 2014 National Graduate Contest on Smart-City Technology, and is one of the winners in the 2014 Bocom Cup Contest on Video Analysis. He is a student member of CCF and IEEE.
  • Supported by:

    This work was supported by the National Natural Science Foundation of China under Grant No. 61173084 and the National Science & Technology Pillar Program of China under Grant No. 2012BAK16B06.

物体轨迹包含丰富的运动信息,因而被广泛应用在高层计算机视觉任务中。考虑到基本跟踪算法简易便捷,大部分高层任务采用的样本是基本跟踪算法获取的原始轨迹,并未显式考虑原始轨迹中的跟踪错误。可靠的轨迹样本是准确建模与识别高层行为的重要前提。因此,本文通过后处理基本跟踪算法的输出轨迹,来修正原始轨迹中的错误样本,提高轨迹可靠性。首先将原始轨迹断裂为短轨迹,再推断并移除存在身份混淆的行人样本,最后通过最大二分图匹配连接短轨迹。本文提出的轨迹后处理框架不依赖于场景先验知识。我们在两个有挑战的数据集做了实验,来验证原始轨迹的修正效果和修正后的轨迹对于高层视觉任务的作用。结果显示,以行为分类为例,修正后的轨迹使得高层任务更加准确,并且修正后的轨迹可以达到与最先进跟踪算法接近的效果。

Abstract: Trajectories carry rich motion cues and thus have been leveraged to many high-level computer vision tasks. Due to the easy implementation of simple trackers, most previous work on trajectory-based applications utilizes raw tracking outputs without explicitly considering tracking errors. Reliable trajectories are prerequisite for modeling and recognizing high-level behaviors. Therefore, this paper tackles such problems by rectifying raw trajectories, which aims to post-process existing trajectories. Our approach firstly splits them into short tracks, and then infers identity ambiguity to remove unquali ed detection responses. At last, short tracks are stitched via maximum bipartite graph matching. This post-processing is completely scene-free. Results of trajectory rectification and their bene ts are both evaluated on two challenging datasets. Results demonstrate that recti ed trajectories are conducive to high-level tasks and the proposed approach is also competitive with state-of-the-art multi-target tracking methods.

[1] Junejo I N, Foroosh H. Trajectory rectification and path modeling for video surveillance. In Proc. the 11th Interna-tional Conference on Computer Vision, October 2007.

[2] Makris D, Ellis T. Learning semantic scene models from observing activity in visual surveillance. IEEE Trans. Sys-tems, Man, and Cybernetics, Part B: Cybernetics, 2005, 35(3): 397-408.

[3] Wang X, Tieu K, Grimson W E L. Correspondence-free activity analysis and scene modeling in multiple camera views. IEEE Trans. Pattern Analysis and Machine Intelligence, 2010, 32(1): 56-71.

[4] Hu W, Li X, Tian G, Maybank S, Zhang Z. An incremental DPMM-based method for trajectory clustering, modeling, and retrieval. IEEE Trans. Pattern Analysis and Machine Intelligence, 2013, 35(5): 1051-1065.

[5] Morris B, Trivedi M. Trajectory learning for activity understanding: Unsupervised, multilevel, and long-term adaptive approach. IEEE Trans. Pattern Analysis and Machine In-telligence, 2011, 33(11): 2287-2301.

[6] Yan W, Forsyth D A. Learning the behavior of users in a public space through video tracking. In Proc. the 7th IEEE Workshops on Application of Computer Vision, January 2005, pp.370-377.

[7] Ma S G, Wang W Q. Effectively discriminating fighting shots in action movies. Journal of Computer Science and Technology, 2011, 26(1): 187-194.

[8] Zhang T, Lu H, Li S Z. Learning semantic scene models by object classification and trajectory clustering. In Proc. IEEE Conference on Computer Vision and Pattern Recog-nition, June 2009, pp.1940-1947.

[9] Chen K W, Lai C C, Lee P J, Chen C S, Hung Y P. Adaptive learning for target tracking and true linking discovering across multiple non-overlapping cameras. IEEE Trans. Multimedia, 2011, 13(4): 625-638.

[10] Javed O, Shafique K, Rasheed Z, Shah M. Modeling intercamera space-time and appearance relationships for tracking across non-overlapping views. Computer Vision and Im-age Understanding, 2008, 109(2):146-162.

[11] Lucas B, Kanade T. An iterative image registration technique with an application to stereo vision. In Proc. IJCAI, Aug. 1981, pp.674-679.

[12] Chen X L, Yang L. Towards monitoring human activities using an omnidirectional camera. In Proc. the 4th IEEE In-ternational Conference on Multimodal Interfaces, October 2002, pp.423-428.

[13] Ferryman J, Shahrokni A. PETS2009: Dataset and challenge. In Proc. the 12th International Workshop on Per-formance Evaluation of Tracking and Surveillance (PETS-Winter), December 2009.

[14] Kuo C H, Huang C, Nevatia R. Multi-target tracking by on-line learned discriminative appearance models. In Proc. IEEE Conference on Computer Vision and Pattern Recog-nition, June 2010, pp.685-692.

[15] Zheng W S, Gong S, Xiang T. Reidentification by relative distance comparison. IEEE Trans. Pattern Analysis and Machine Intelligence, 2013, 35(3): 653-668.

[16] Yang B, Nevatia R. Online learned discriminative partbased appearance models for multi-human tracking. In Proc. the 12th European Conference on Computer Vision, October 2012, pp.484-498.

[17] Huang C, Wu B, Nevatia R. Robust object tracking by hierarchical association of detection responses. In Proc. the 10th European Conference on Computer Vision, October 2008, pp.788-801.

[18] Milan A, Schindler K, Roth S. Detection-and trajectorylevel exclusion in multiple object tracking. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, June 2013, pp.3682-3689.

[19] Henriques J F, Caseiro R, Batista J. Globally optimal solution to multi-object tracking with merged measurements. In Proc. IEEE International Conference on Computer Vision, November 2011, pp.2470-2477.

[20] Zamir A R, Dehghan A, Shah M. GMCP-tracker: Global multi-object tracking using generalized minimum clique graphs. In Proc. the 12th European Conference on Com-puter Vision, October 2012, pp.343-356.

[21] Xu Y, Qin L, Li G, Huang Q. Online discriminative structured output SVM learning for multi-target tracking. IEEE Signal Processing Letters, 2014, 21(2): 190-194.

[22] Andriyenko A, Schindler K. Multi-target tracking by continuous energy minimization. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, June 2011, pp.1265-1272.

[23] Wang B, Wang G, Chan K L, Wang L. Tracklet association with online target-specific metric learning. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, June 2014, pp.1234-1241.

[24] Yang B, Nevatia R. Multi-target tracking by online learning a CRF model of appearance and motion patterns. Interna-tional Journal of Computer Vision, 2014, 107(2): 203-217.

[25] Pirsiavash H, Ramanan D, Fowlkes C C. Globally-optimal greedy algorithms for tracking a variable number of objects. In Proc. IEEE Conference on Computer Vision and Pat-tern Recognition, June 2011, pp.1201-1208.

[26] Berclaz J, Fleuret F, Turetken E, Fua P. Multiple object tracking using k-shortest paths optimization. IEEE Trans. Pattern Analysis and Machine Intelligence, 2011, 33(9): 1806-1819.

[27] Yu Q, Medioni G. Multiple-target tracking by spatiotemporal Monte Carlo Markov chain data association. IEEE Trans. Pattern Analysis and Machine Intelligence, 2009, 31(12): 2196-2210.

[28] Rasmussen C, Hager G D. Probabilistic data association methods for tracking complex visual objects. IEEE Trans. Pattern Analysis and Machine Intelligence, 2001, 23(6): 560-576.

[29] Guo C C, Chen S Z, Lai J H, Hu X J, Shi S C. Multi-shot person re-identification with automatic ambiguity inference and removal. In Proc. International Conference on Pattern Recognition, August 2014, pp.3540-3545.

[30] Johnson S C. Hierarchical clustering schemes. Psychome-trika, 1967, 32(3): 241-254.

[31] Belongie S, Malik J, Puzicha J. Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Analysis and Machine Intelligence, 2002, 24(4): 509-522.

[32] Lian G, Lai J H, Zheng W S. Spatial-temporal consistent labeling of tracked pedestrians across non-overlapping camera views. Pattern Recognition, 2011, 44(5): 1121-1136. 372 J. Comput. Sci. & Technol., Mar. 2015, Vol.30, No.2

[33] Cheung Y. Rival penalization controlled competitive learning for data clustering with unknown cluster number. In Proc. the 9th International Conference on Neural Informa-tion Processing, November 2002, pp.467-471.

[34] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In Proc. IEEE Conference on Computer Vi-sion and Pattern Recognition, June 2005, pp.886-893.

[35] Zivkovic Z. Improved adaptive Gaussian mixture model for background subtraction. In Proc. the 17th International Conference on Pattern Recognition, August 2004, pp.28-31.

[36] Hare S, Saffari A, Torr P. Struck: Structured output tracking with kernels. In Proc. IEEE International Conference on Computer Vision, November 2011, pp.263-270.

[37] Kalal Z, Mikolajczyk K, Matas J. Tracking-learningdetection. IEEE Trans. Pattern Analysis and Machine In-telligence, 2012, 34(7): 1409-1422.
No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 金凌紫;. TrapML-A Metalanguage for Transformational Programming[J]. , 1990, 5(4): 388 -399 .
[2] 周权; 魏道政;. A Complete Critical Path Algorithm for Test Generation of Combinational Circuits[J]. , 1991, 6(1): 74 -82 .
[3] 熊志国; 徐曦; 董士海;. CX11: A Chinese Language Supporting Interface for X Window Environment[J]. , 1995, 10(1): 15 -22 .
[4] 沈一栋;. A General Scheme for Formalizing Defaults Usingthe Predicate ab(I,S)[J]. , 1999, 14(2): 159 -164 .
[5] . 半封闭数据立方体:一种有效平衡数据立方体体积和查询响应时间的实现方法[J]. , 2005, 20(3): 367 -372 .
[6] . 基于边缘纹理驱动模型的贝叶斯人脸定位[J]. , 2005, 20(6): 849 -854 .
[7] . 暂缺[J]. , 2006, 21(1): 19 -26 .
[8] . AVS-M:从标准到应用[J]. , 2006, 21(3): 332 -344 .
[9] . 暂缺[J]. , 2006, 21(6): 989 -996 .
[10] . KASUM算法的更高水平的硬件综合实现方法[J]. , 2007, 22(1): 60 -70 .
版权所有 © 《计算机科学技术学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn
总访问量: