期刊文献+

动作识别算法的评估策略探讨 被引量:10

Discussion on the assessment strategy of action recognition algorithms
原文传递
导出
摘要 以时空兴趣点特征和支持向量机(SVM)分类器识别方法为基本算法,在广泛使用的公开动作数据集KTH上,从不同角度考察评估策略对动作识别算法性能的影响。实验表明,当采用不同的交叉实验方法时,算法性能的波动最大达到10.5%,而不同数据集划分方法对算法性能的影响则达到11.87%。因此,通过量化分析得出的结论,可以充分地比较现有算法的真实差异,并为设计合理的评估策略提供参考。 Action recognition is a hot research topic,but the performance assessment strategies of algorithms have not had an accepted practice.In this paper,we adopt spatio-temporal features and support vector machine(SVM) model as our action recognition algorithm,and then well assess the effect of different assessment strategies on our action recognition algorithm in widely used public dataset KTH.Experimental results show that when different cross-experimental methods are employed,the performance fluctuation of algorithms reaches 10.5%.And when different division methods for KTH datasets are used,the performance fluctuation of algorithms gets 11.87%.Thus,according to conclusions in this paper,we can find the real difference among existing algorithms,and supply the reference for designing reasonable assessment strategy.
出处 《光电子.激光》 EI CAS CSCD 北大核心 2012年第6期1166-1172,共7页 Journal of Optoelectronics·Laser
基金 国家自然科学基金(90920001) 天津市科技支撑计划(10ZCKFGX00400)资助项目
关键词 时空特征 支持向量机(SVM) 动作识别 评估策略 spatio-temporal features support vector machine(SVM) action recognition assessment strategy
  • 相关文献

参考文献1

二级参考文献16

  • 1王典,程咏梅,杨涛,潘泉,赵春晖.基于混合高斯模型的运动阴影抑制算法[J].计算机应用,2006,26(5):1021-1023. 被引量:20
  • 2杜友田,陈峰,徐文立,李永彬.基于视觉的人的运动识别综述[J].电子学报,2007,35(1):84-90. 被引量:79
  • 3Aggarwal J. K, Cai Q. Human motion analysis:A review[J]. Computer Vision and Image Understanding, 1999, 73 ( 3 ) : 428- 440.
  • 4Gavrila D. M. The visual analysis of human movement :A survey [J]. Computer Vision and Image Understanding, 1999,73 ( 1 ) : 82-98.
  • 5Bashir F I, Khokhar A A, Schonfeld D. Object trajectory based activity classification and recognition using hidden Markov models[J]. IEEE Trans on Image Processing, 2007,16 (7) : 1912-1919.
  • 6ZHU Guang-yu, XU Chang-sheng. Action recognition in broadcast tennis video[A]. Proc of the 18th International Conference on Pattern Recognition[C]. N J: IEEE,2006. 251-254.
  • 7Chert H S,Chen H T,Chen Y W,et al. Human action recognition using star skeleton [A]. Proc of the 4th ACM International Workshop on Video Surveillance and Sensor Networks[C]. NY:AOM,2006. 179-]82.
  • 8HOU Ye, GUO Bao-long Human motion detection by using graph cuts[J]. Journal of Optoelectronics · Laser, 2007, 18 (6) :725-728.
  • 9Salmon J P, Debled-Rennesson I, Wendling L. A new method to detect arcs and segments from curvature profiles[A]. Proc of the 18th International Conference on Pattern Recognition (ICPR'06) [C]. IEEE,2006. 387-390.
  • 10Fujiyoshi H,Lipton A. Real-time human motion analysis by image skeletonization[A]. In, Proceedings of IEEE Workshop on Applications of Computer Vision[C]. NJ:IEEE, 1998. 15-21.

共引文献21

同被引文献133

  • 1Dollar P,Rabaud V,Cottrell G. et al. Behavior recognition via sparse spatio-temporal features[A]. IEEE Internation- al Workshop on Visual Surveillance and Performance E- valuation of Tracking and Surveillance[C]. 2005,65-72.
  • 2Laptev I, Lindeberg T. Space-time interest points[A]. Proc. of IOOV, [C]. 2003,432-439.
  • 3Liu J, Shah M. Learning human actions via information maximization[A]. Proc. of CVPR[C]. 2008,1-8.
  • 4Bobick A F, Davis J W. The recognition of human move- ment using temporal templates [J]. PAMI, 2001,23 ( 3 ): 257-267.
  • 5Kellokumpu V, Pietikainen M, Heikkila J. Human activity recognition using sequences of postures[A]. Proc. of IA- PR[C]. 2005,570-573.
  • 6Zhang J,Gong S. Action categorization with modified hid- den conditional random field [J]. Pattern Recognition, 2010,43 : 197-203.
  • 7Wang J,Liu Z. Mining actionlet ensemble for action rec- ognition with depth cameras [A]. Proc. of CVPR [C]. 2012,1290-1297.
  • 8Li W,Zhang Z,Liu Z. Action recognition based on a bag of 3D points[A]. Proc. of CVPR[C]. 2010,9-14.
  • 9Megavannan V, Agarwal B. Human action recognition u- sing depth maps[A]. Proc. of SPCOM[C]. 2012, 1-5.
  • 10Lin Y C,Hu M C. Human action recognition and retrieval using sole depth information[A]. Proc. of ACM[C]. 2012, 1053-1056.

引证文献10

二级引证文献70

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部