Action Recognition Using Graph Embedding and the Co-occurrence Matrices Descriptor

Feng Zheng (1) , Ling Shao (2) , Zhan Song (1) , Xi Chen (1)

1 SIAT - Shenzhen Institute of Advanced Technology [Shenzhen]
2 Electronic and Electrical Engineering

Recognizing actions from a monocular video is a very hot topic in computer vision recently. In this paper, we propose a new representation of actions, the co occurrence matrices de-scriptor, on the intrinsic shape manifold learned by graph embedding. The co-occurrencematrices descriptor captures more temporal information than the bag of words (histogram) descriptor which only considers the spatial information, thus boosts the classi¯cation accuracy. In addition, we compare the performance of the co-occurrence matrices descriptor on different manifolds learned by various graph embedding methods. Graph embedding methods preserve as much of the signi¯cant structure of the high-dimensional data as possible in the low-dimensional map. The results show that nonlinear algorithms are more robust than linear ones. Furthermore, we conclude that the label information plays a critical role in learning more discriminating manifolds.

Mots clés: Physical Sciences

Domaines: Informatique [cs]

Feng Zheng, Ling Shao, Zhan Song, Xi Chen. Action Recognition Using Graph Embedding and the Co-occurrence Matrices Descriptor. International Journal of Computer Mathematics, 2011, ⟨10.1080/00207160.2011.578741⟩. ⟨hal-00711298⟩