Small Group Human Activity Recognition

Small human group activity recognition has attracted much attention in recent years, since human activities often represent as small groups in public surveillance systems. Comparing to single human activity recognition or crowd analysis, small human group activity recognition is much more challenging due to mutual occlusions between different people, the varying group size, and inter or intra group interactions. 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 Maximum Gaussian Process Dynamical Model(MGPDM) is then employed to learn the temporal activity of small group people overtime. After training, the testing group activity will be identified as the action with the highest conditional probability with respect to each trained activity model. Our experimental results indicate that the proposed features and behavior model can successfully capture both the spatial and temporal dynamics of group people behavior, and correctly identify different small human group activities.