| 自然环境下日常动作的在线识别 |
| |
| 引用本文： | 曹媛媛,黄飞跃,陶霖密,徐光祐. 自然环境下日常动作的在线识别[J]. 电子学报, 2009, 37(Z1): 16-21 |
| |
| 作者姓名： | 曹媛媛 黄飞跃 陶霖密 徐光祐 |
| |
| 作者单位： | 1. 普适计算教育部重点实验室, 清华大学计算机系, 北京, 100084;2. 腾讯研究院, 北京, 100080 |
| |
| 摘 要： | 自然环境下的日常动作识别有着广泛的应用前景和重要的研究价值.不同于以往在结构化和孤立条件下进行的动作识别,自然环境下的日常动作是连续的,视角多变并常发生遮挡.本文提出了分布式视觉系统下日常动作的在线识别方法.时间轴上的滑动窗口每个时刻取一段视频帧,采用基于"包容形状"的视角无关的体态表示方法提取体态特征向量,并用隐马尔科夫模型进行识别.动作类型的搜索空间由环境知识推理得到.遮挡检测和部分遮挡下的体态表示也在文中进行了讨论.实验表明本文提出的日常动作的在线识别方法能够克服日常场景给动作识别带来的困难,结果证实了方法的有效性.
|
| 关 键 词： | 动作识别 视角不变 遮挡检测 |
| 收稿时间： | 2008-11-06 |
On-Line Recognition of Actions in Daily Living |
| |
| CAO Yuan-yuan,HUANG Fei-yue,TAO Lin-mi,XU Guang-you. On-Line Recognition of Actions in Daily Living[J]. Acta Electronica Sinica, 2009, 37(Z1): 16-21 |
| |
| Authors: | CAO Yuan-yuan HUANG Fei-yue TAO Lin-mi XU Guang-you |
| |
| Affiliation: | 1. Key Laboratory of Pervasive Computing of Ministry of Education, Department of ComputerScience and Technology, Tsinghua University, Beijing 100084;2. Tencent Research, Beijing 100080 |
| |
| Abstract: | Recognition of actions in daily living is challenging because:(1) actions are continuous;(2) human location is changeful;(3) human body is partially occluded sometimes.This paper proposes a multi-view framework for on-line recognition of actions in daily living.Action representation based on "Envelope shape" enables view-invariant action recognition.A sliding window concatenates the feature vectors for action representation into a stream as the input to a bank of HMMs.A maximum likelihood based classifier d... |
| |
| Keywords: | action recognition view invariant occlusion detection |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《电子学报》浏览原始摘要信息 |
|
点击此处可从《电子学报》下载免费的PDF全文 |
|
