2012 IEEE Conference on Computer Vision and Pattern Recognition (2012)
Providence, RI USA
June 16, 2012 to June 21, 2012
G. Mori , Simon Fraser Univ., Burnaby, BC, Canada
L. Sigal , Disney Res., Pittsburgh, PA, USA
Tian Lan , Simon Fraser Univ., Burnaby, BC, Canada
We present a hierarchical model for human activity recognition in entire multi-person scenes. Our model describes human behaviour at multiple levels of detail, ranging from low-level actions through to high-level events. We also include a model of social roles, the expected behaviours of certain people, or groups of people, in a scene. The hierarchical model includes these varied representations, and various forms of interactions between people present in a scene. The model is trained in a discriminative max-margin framework. Experimental results demonstrate that this model can improve performance at all considered levels of detail, on two challenging datasets.
video signal processing, behavioural sciences computing, image recognition, video event, social role, hierarchical model, human activity recognition, multiperson scene, human behaviour, low-level action, high-level event, discriminative max-margin framework, event recognition, Humans, Support vector machines, Video sequences, Context, Vectors, Context modeling, Surveillance
G. Mori, L. Sigal and Tian Lan, "Social roles in hierarchical models for human activity recognition," 2012 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), Providence, RI USA, 2012, pp. 1354-1361.