Human body pose estimation using silhouette shape analysis.pdfVIP

下载本文档

10
0
约2.92万字
约 8页
2017-04-11 发布于江苏
举报
版权申诉

Human body pose estimation using silhouette shape analysis.pdf

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Human body pose estimation using silhouette shape analysis

Human Body Pose Estimation Using Silhouette Shape Analysis Anurag Mittal ? Liang Zhao, Larry S. Davis Siemens Corporate Research Inc. Department of Computer Science Princeton, NJ 08540 University of Maryland College Park, MD 20742 Abstract We describe a system for human body pose estimation from multiple views that is fast and completely automatic. The algorithm is able to work in the presence of multiple people by decoupling the problems of pose estimation of different people.The pose is estimated based on a likelihood function that integrates information from multiple views and thus ob- tains a globally optimum solution. Other characteristics that make our method more general than previous work in- clude: (1) no manual initialization, (2) no specification of the dimensions of the 3D structure, (3) no reliance on some learned poses or patterns of activity, and (4) insensitivity to edges and clutter in the background and within the fore- ground. The algorithm has applications in surveillance and promising results have been obtained. 1 Introduction Determining the pose of humans is an important problem in vision and has many applications. We describe a system for human body pose estimation from multiple views that is suitable for surveillance applications. The emphasis is on building a system that is fast and general, rather than on very accurate recovery of the pose parameters under constrained scenarios. Our algorithm has the characteristic of being completely automatic. Manual input in prior work typically takes the form of either initialization of pose parameters or careful specification of the dimensions of the articulated structure. Another characteristic that limits the applicability of prior work is the sensitivity to edges in the scene, both in the background and within the foreground. Since our algorithm relies solely on occluding contours, it is completely insensi- tive to both internal and background edges. Another charac- teristic that lends more general