TY - GEN
T1 - Small group human activity recognition
AU - Yin, Yafeng
AU - Yang, Guang
AU - Xu, Jin
AU - Man, Hong
PY - 2012
Y1 - 2012
N2 - Small group people activity recognition has attracted much attention in computer vision community in recent years, since it has great potential in public security applications. Comparing to single human activity recognition, group human activity recognition has much more challenges, such as mutual occlusions between different people, the varying group size, and the interaction within or between groups. In this paper, we propose a novel structural feature set to represent group behavior as well as a probabilistic framework for group activity learning and recognition. We first apply a robust multiple targets tracking algorithm to track each individual in the entire image region. Small groups are then clustered based on the output positions of the tracker. After that, we introduce a set of social network analysis based structural features to describe the dynamic behavior of small group people in each frame. A Gaussian Process Dynamical Model(GPDM) is then employed to learn the temporal activity of small group people overtime. After training, the new group activity will be identified by computing the conditional probability with each learned GPDM. Our experimental results indicate that our proposed features and behavior model can successfully capture both the spatial and temporal dynamics of group people behavior, and correctly identify different group activities.
AB - Small group people activity recognition has attracted much attention in computer vision community in recent years, since it has great potential in public security applications. Comparing to single human activity recognition, group human activity recognition has much more challenges, such as mutual occlusions between different people, the varying group size, and the interaction within or between groups. In this paper, we propose a novel structural feature set to represent group behavior as well as a probabilistic framework for group activity learning and recognition. We first apply a robust multiple targets tracking algorithm to track each individual in the entire image region. Small groups are then clustered based on the output positions of the tracker. After that, we introduce a set of social network analysis based structural features to describe the dynamic behavior of small group people in each frame. A Gaussian Process Dynamical Model(GPDM) is then employed to learn the temporal activity of small group people overtime. After training, the new group activity will be identified by computing the conditional probability with each learned GPDM. Our experimental results indicate that our proposed features and behavior model can successfully capture both the spatial and temporal dynamics of group people behavior, and correctly identify different group activities.
KW - GPDM
KW - human group action recognition
KW - social network features
UR - http://www.scopus.com/inward/record.url?scp=84875862342&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84875862342&partnerID=8YFLogxK
U2 - 10.1109/ICIP.2012.6467458
DO - 10.1109/ICIP.2012.6467458
M3 - Conference contribution
AN - SCOPUS:84875862342
SN - 9781467325332
T3 - Proceedings - International Conference on Image Processing, ICIP
SP - 2709
EP - 2712
BT - 2012 IEEE International Conference on Image Processing, ICIP 2012 - Proceedings
T2 - 2012 19th IEEE International Conference on Image Processing, ICIP 2012
Y2 - 30 September 2012 through 3 October 2012
ER -