| Human action recognition from realistic videos is a challenging problem in computer vision. Several intrinsic properties such as intra-class variations, background clutter and partial occlusion make it difficult to recognize individual person actions reliably.;In this dissertation, we go beyond recognizing individual person actions and focus on group activities instead. This motivates from the observation that human actions are rarely performed in isolation, the contextual information of what other people nearby are doing provides useful cues for understanding the high-level activities. We propose a discriminative model for recognizing group activities. Our model jointly captures the group activity, the individual person actions, and the interactions among them. Two new types of contex- tual information, group-person interaction and person-person interaction, are explored in a latent variable framework. In particular, we propose two different approaches to model the person-person interaction. One approach is to explore the structures of person-person interaction. Different from most of the previous latent structured models which assume a pre-defined structure for the hidden layer, e.g. a tree structure, we treat the structure of the hidden layer as a latent variable and implicitly infer it during learning and inference. The other approach explores the person-person interaction in feature level. We introduce a new feature representation called the action context (AC) descriptor. The AC descriptor encodes information about not only the action of an individual person in the video, but also the behaviour of other people nearby. Our experimental results demonstrate the benefit of using contextual information for disambiguating group activities. |
