| 引用本文: | 谭论正,夏利民,黄金霞,等.基于pLSA模型的人体动作识别.[J].国防科技大学学报,2013,35(5):102-108.[点击复制] |
| TAN Lunzheng,XIA Limin,HUANG Jinxia,et al.Human action recognition based on pLSA model[J].Journal of National University of Defense Technology,2013,35(5):102-108[点击复制] |
|
| |
|
|
| 本文已被：浏览 7639次 下载 7387次 |
| 基于pLSA模型的人体动作识别 |
| 谭论正1, 夏利民1, 黄金霞1, 夏胜平2 |
|
(1.中南大学 信息科学与工程学院，湖南 长沙 410075;2.国防科技大学 ATR重点实验室，湖南 长沙 410073)
|
| 摘要: |
| 提出一种基于主题模型的人体动作识别方法。该方法首先提取时空兴趣点(STIP，Space-Time Interest Point)来描述人体运动，然后提出使用慢特征分析(SFA, Slow Feature Analysis)算法计算兴趣点梯度信息不变量最优解，最后使用概率潜在语义分析(pLSA, probabilistic Latent Semantic Analysis) 模型识别人体动作。SFA计算的梯度不变量最优解可以表示时空兴趣点固有特征，能够无歧义反映时空兴趣点在空间及时间方向上的信息。同时，针对pLSA隐性主题正确性无法保证的缺点，算法将主题与动作标签“一对一”相关，通过监督方式得到主题，保证了训练中主题的正确性。该算法在KTH人体运动数据库和Weizmann人体动作数据库进行了训练与测试，动作识别结果正确率分别在91.50%和97%以上。 |
| 关键词: 动作识别 主题模型 慢特征分析 时空兴趣点 梯度直方图 |
| DOI： |
| 投稿日期：2013-05-28 |
| 基金项目:国家863高技术资助项目(2009AA11Z205)；国家自然科学基金资助项目(50808025)；教育部博士点基金资助项目(20090162110057) |
|
| Human action recognition based on pLSA model |
| TAN Lunzheng1, XIA Limin1, HUANG Jinxia1, XIA ShengPing2 |
|
(1.College of Information Science and Engineering，Central South University ,Changsha 410075，China;2.ATR State Key Lab, National University of Defense Technology, Changsha 410073, China.)
|
| Abstract: |
| A human action recognition method based on a probabilistic topic model is proposed. Firstly, the method extracts space-time interest points to describe human motion. Then the slow feature analysis algorithm was proposed to calculate the invariant optimal solution of the gradient information of space time points. Lastly human actions were recognized with the probabilistic latent semantic analysis(pLSA). The invariant optimal solution of the gradient information can express the inherent characteristics of STIP, and it can also reflect the space and time information of STIP discriminatively. For solving the problem of latent topics that are not guaranteed in pLSA, the topics obtained in supervised fashion correspond to action labels one by one. Action recognition results were presented on KTH human motion data set and Weizmann human action data set. Our results show that the action recognition rates of the tow dataset are respectively more than 91.50% and 97%. |
| Keywords: action recognition topic model slow feature analysis space-time interest points histogram of gradient |
|
|
|
|
|
