Graphs are a powerful tool to model structured objects, but it is nontrivial to measure the similarity between two graphs. In this paper, we construct a two-graph model to represent human actions by recording the spatial and temporal relationships among local features. We also propose a novel family of context-dependent graph kernels (CGKs) to measure similarity between graphs. First, local features are used as the vertices of the two-graph model and the relationships among local features in the intra-frames and inter-frames are characterized by the edges. Then, the proposed CGKs are applied to measure the similarity between actions represented by the two-graph model. Graphs can be decomposed into numbers of primary walk groups with different walk lengths and our CGKs are based on the context-dependent primary walk group matching. Taking advantage of the context information makes the correctly matched primary walk groups dominate in the CGKs and improves the performance of similarity measurement between graphs. Finally, a generalized multiple kernel learning algorithm with a proposed l12-norm regularization is applied to combine these CGKs optimally together and simultaneously train a set of action classifiers. We conduct a series of experiments on several public action datasets. Our approach achieves a comparable performance to the stateof-the-art approaches, which demonstrates the effectiveness of the two-graph model and the CGKs in recognizing human actions.

In this paper, a multi-feature max-margin hierarchical Bayesian model (M 3 HBM) is proposed for action recognition. Different from existing methods which separate representation and classification into two steps, M 3 HBM jointly learns a high-level representation by combining a hierarchical generative model (HGM) and discriminative maxmargin classifiers in a unified Bayesian framework. Specifically, HGM is proposed to represent actions by distributions over latent spatial temporal patterns (STPs) which are learned from multiple feature modalities and shared among different classes. For recognition, we employ Gibbs classifiers to minimize the expected loss function based on the max-margin principle and use the classifiers as regularization terms of M 3 HBM to perform Bayeisan estimation for classifier parameters together with the learning of STPs. In addition, multi-task learning is applied to learn the model from multiple feature modalities for different classes. For test videos, we obtain the representations by the inference process and perform action recognition by the learned Gibbs classifiers. For the learning and inference process, we derive an efficient Gibbs sampling algorithm to solve the proposed M 3 HBM. Extensive experiments on several datasets demonstrate both the representation power and the classification capability of our approach for action recognition.

The performance of action recognition in video sequences depends significantly on the representation of actions and the similarity measurement between the representations. In this paper, we combine two kinds of features extracted from the spatio-temporal interest points with context-aware kernels for action recognition. For the action representation, local cuboid features extracted around interest points are very popular using a Bag of Visual Words (BOVW) model. Such representations, however, ignore potentially valuable information about the global spatio-temporal distribution of interest points. We propose a new global feature to capture the detailed geometrical distribution of interest points. It is calculated by using the 3D R transform which is defined as an extended 3D discrete Radon transform, followed by the application of a two-directional two-dimensional principal component analysis. For the similarity measurement, we model a video set as an optimized probabilistic hypergraph and propose a context-aware kernel to measure high order relationships among videos. The context-aware kernel is more robust to the noise and outliers in the data than the traditional context-free kernel which just considers the pairwise relationships between videos. The hyperedges of the hypergraph are constructed based on a learnt Mahalanobis distance metric. Any disturbing information from other classes is excluded from each hyperedge. Finally, a multiple kernel learning algorithm is designed by integrating the l 2 norm regularization into a linear SVM classifier to fuse the R feature and the BOVW representation for action recognition. Experimental results on several datasets demonstrate the effectiveness of the proposed approach for action recognition.

Graphs are effective tools for modeling complex data. Setting out from two basic substructures, random walks and trees, we propose a new family of context-dependent random walk graph kernels and a new family of tree pattern graph matching kernels. In our context-dependent graph kernels, context information is incorporated into primary random walk groups. A multiple kernel learning algorithm with a proposed l1,2-norm regularization is applied to combine context-dependent graph kernels of different orders. This improves the similarity measurement between graphs. In our tree-pattern graph matching kernel, a quadratic optimization with a sparse constraint is proposed to select the correctly matched tree-pattern groups. This augments the discriminative power of the tree-pattern graph matching. We apply the proposed kernels to human action recognition, where each action is represented by two graphs which record the spatiotemporal relations between local feature vectors. Experimental comparisons with state-of-the-art algorithms on several benchmark datasets demonstrate the effectiveness of the proposed kernels for recognizing human actions. It is shown that our kernel based on tree-pattern groups, which have more complex structures and exploit more local topologies of graphs than random walks, yields more accurate results but requires more runtime than the context-dependent walk graph kernel.