A View-based Multiple Objects Tracking and Human Action Recognition for Interactive Virtual Environments
As environments become smart in accordance withadvances in ubiquitous computing technology, researchers are struggling to satisfy users diverse and sophisticated demands.The aim of the present work is to enable multiple persons in ainteractive virtual environment to simultaneously andconveniently interact with virtual agents. To this end, we proposea real-time system that robustly tracks multiple persons in virtualenvironments and recognizes their actions through imagesequences acquired from a single fixed camera. The proposed system is compromised of three components: blob extraction,object tracking, and human action recognition. Given an image,we extract blobs using the Mixture of Gaussians algorithm with a hierarchical data structure and we additionally remove shadowsand highlights in order to obtain a more accurate object silhouette.We then track multiple objects using a motion-based object modeland an inference graph for handling grouping and fragmentproblems. Finally, we model an action as a Motion History Image(MHI) based on given object tracks, normalize the MHI, reducethe MHI using PCA, and classify an action using a multi-layerperceptron. To evaluate the performance of the proposed system,we employed it in an augmented reality application where multiple persons can interact with a virtual pet. The results confirm that reliable object tracking is achieved and multiple persons actions can be recognized for applications in interactive virtual environments.
Human action recognition human computer interaction object detection object tracking
Jin Choi Yong-il Cho Kyusung Cho Suojung Bae Hyun S. Yang
AIM Lab.,Dept.of Electrical Engineering and Computer Science,KAIST,Daejeon,South Korea
国际会议
2008年国际虚拟环境和虚拟智能大会(IVEVA 2008)
南京
英文
163-168
2008-06-01（万方平台首次上网日期，不代表论文的发表时间）