期刊文献+

基于概率主题模型的动作识别算法研究

A novel method of action recognition based on probabilistic topic model
下载PDF
导出
摘要 在时空兴趣点以及3D-SIFT描述的基础上,采用概率主题模型进行动作识别,结果表明:概率主题模型不仅能够实现对视频中单个动作的识别,而且对复杂情况下的视频也有一定识别能力。 A novel method of action recognition based on probabilistic topic model is proposed to recognize several human common actions,on the basis of spatial-temporal interest point and 3D-SIFT description.The result shows that the probabilistic topic model has the recognition not only to single action in video,but also to the sequential video action under complex condition.
作者 王向慧
出处 《长春工程学院学报(自然科学版)》 2011年第3期113-117,共5页 Journal of Changchun Institute of Technology:Natural Sciences Edition
关键词 人体行为识别 时空兴趣点 主题模型 human action recognition spatial-temporal interest point topic models
  • 相关文献

参考文献9

  • 1Niebles J C,Wang H,Fei-Fe L i.Unsupervised learningof human action categories using spatial-temporalwords[J].International Journal of Computer Vision,2008,79(3):299-318.
  • 2Blei,D M,Ng,etc.Latent Dirichlet allocation[J].Journalof Machine Learning Research,2003,13(3):993-1022.
  • 3Yang Wang,Greg Mori.Human action recognition bysemilatent topic models[J].IEEE Transactions on Pat-tern Analysis and Machine Intelligence(PAMI),2009,31(10):1762-1774.
  • 4Scovanner P,Ali S,Shah M.A 3-dimensional sift de-scriptor and its application to action recognition[A].Shah M.In Proceedings of the 15th international confer-ence on Multimedia[C].Beijing,2007:357-360.
  • 5Dollar P,Rabaud V,Cottrell G,etc.Behavior recognitionvia sparse spatio-temporal features[J].IEEE Interna-tional Workshop on Visual Surveillance and PerformanceEvaluation of Tracking and Surveillance,2005,16(3):65-72.
  • 6BLANKM,OREL I CK L G,SHECHT MAN E,etc.Ac-tions as space time shapes[A].SHECHT MANE.Procof IEEE International Conference on Computer Vision[C].Beijing:2005:167-172.
  • 7Schuldt,Laptev C,Caputo B I.Recognizing human actions:alocal SVM approach[J].ICPR,2004,17(2):32-36.
  • 8Dhillon P S,Nowozin S,Lampert C H..Combining ap-pearance and motion for human action classification invideo[J].CVPR,2008,23(1):1-8.
  • 9Liu J,Ali S,Shah M.Recognizing human actions usingmultiple features[J].CVPR,2008,23(3):1-8.

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