TY - GEN
T1 - Human action recognition using star templates and delaunay triangulation
AU - Chuang, Chi Hung
AU - Hsieh, Jun Wei
AU - Tsai, Luo Wei
AU - Fan, Kuo Chin
PY - 2008
Y1 - 2008
N2 - This paper presents a human action recognition system for recognizing various behaviors directly from videos. Firstly, we triangulate the human body to different triangle meshes. Then, we use a depth-first search (dfs) scheme to find a spanning tree from the set of meshes. All leafs of the spanning tree are adopted as the extremities. Different from traditional approaches to find the extremities on the target's silhouette as skeletons, the extremities found from the internal centroids of triangle meshes can represent a human posture more accurately and robustly. To model each human action, all the input skeleton sequences are then transformed into symbol sequences. Then, we design a string matching scheme to measure the similarity between any two human behaviors. Since 2D postures are used in this paper, the above scheme is sensitive to different view points. To solve the view independent problem, a 2D matrix is then constructed for recording the symbol relations between two viewpoints. Thus, our proposed matching scheme is almost view-invariant. Experimental results show that the proposed scheme is a robust, efficient, and promising tool in human action recognition.
AB - This paper presents a human action recognition system for recognizing various behaviors directly from videos. Firstly, we triangulate the human body to different triangle meshes. Then, we use a depth-first search (dfs) scheme to find a spanning tree from the set of meshes. All leafs of the spanning tree are adopted as the extremities. Different from traditional approaches to find the extremities on the target's silhouette as skeletons, the extremities found from the internal centroids of triangle meshes can represent a human posture more accurately and robustly. To model each human action, all the input skeleton sequences are then transformed into symbol sequences. Then, we design a string matching scheme to measure the similarity between any two human behaviors. Since 2D postures are used in this paper, the above scheme is sensitive to different view points. To solve the view independent problem, a 2D matrix is then constructed for recording the symbol relations between two viewpoints. Thus, our proposed matching scheme is almost view-invariant. Experimental results show that the proposed scheme is a robust, efficient, and promising tool in human action recognition.
UR - http://www.scopus.com/inward/record.url?scp=54049130494&partnerID=8YFLogxK
U2 - 10.1109/IIH-MSP.2008.342
DO - 10.1109/IIH-MSP.2008.342
M3 - 會議論文篇章
AN - SCOPUS:54049130494
SN - 9780769532783
T3 - Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008
SP - 179
EP - 182
BT - Proceedings - 2008 4th International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP 2008
T2 - 2008 4th International Conference on Intelligent Information Hiding and Multiedia Signal Processing, IIH-MSP 2008
Y2 - 15 August 2008 through 17 August 2008
ER -