Gesture Recognition Based On Elastic Deformation Energies.pdf

Gesture Recognition Based On Elastic Deformation Energies.pdf

  1. 1、本文档共12页,可阅读全部内容。
  2. 2、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。
  3. 3、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
  4. 4、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)。
查看更多
Gesture Recognition Based On Elastic Deformation Energies

Gesture Recognition Based On Elastic Deformation Energies Radu-Daniel Vatavu1,2, Laurent Grisoni1, and Stefan-Gheorghe Pentiuc2 1 Laboratoire d’Informatique Fondamentale de Lille, Villeneuve d’Ascq 59650, France 2 University Stefan cel Mare, Suceava 720229, Romania radu.vatavu@lifl.fr,laurent.grisoni@lifl.fr,pentiuc@eed.usv.ro Abstract. We present a gesture recognition method based on deformable shapes and curvature templates. Gestures are modeled using a spline rep- resentation that is enhanced with elastic properties: a gesture trajectory as a whole or any of its parts may stretch or bend. We regard such an ap- proach as well-suited for dealing with the inherent variability of human gesture execution. The results of our gesture classifier are demonstrated with a video-based acquisition approach. Key words: gesture recognition, elastic matching, deformation energies 1 Introduction Gesture has been given a lot of attention in the last decades as an effective mean for human computer interaction. The motivation is given by naturalness and efficiency: people use gestures in the real world to interact with real objects and to convey information. An interface that is based on the use of gestures may prove to be ideal comparing to other interaction techniques. That is of course if we consider an ideal gesture framework that does not importunate, distract or adds significant cognitive load [1, 2]. Successful implementation of a gesture recognition system includes selection of an appropriate technology for acquisition, defining a model for gesture rep- resentation, implementing a robust classifier and finally, giving feedback to the user. Many of the above issues have been widely discussed in the literature. Good overviews on the state of the art in gesture interaction including gesture taxonomies for HCI, existing technologies, recognition and interpretation tech- niques are given in [3–5]. A particular problem that may be identified among the ones listed above relates to

文档评论(0)

l215322 + 关注
实名认证
内容提供者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档