A Tensor-Driven Temporal Correlation Model For Video Sequence Classification

With a dramatic increasing number of video data,automatic video-based classification has aroused growing attention in video analysis.The video sequence classification has been widely used in various fields including video summarization,video retrieval,and action recognition.Human actions and gestures are the most key semantic components of video data when classifying the video sequence.Meanwhile the task of video sequence classification plays a critical role in the development of computer vision.Considering this fact,this paper proposes a novel tensor decomposition method called Tensor-Driven Temporal Correlation,in which general tensors are used as input for video sequence classification.Because distortion and redundancy may exist in the tensor representations of video sequences,we project the original tensor into subspaces spanned by spatial basis matrices in the proposed formulation.Moreover,to better preserve the temporal smoothness between consecutive slices of the tensor,the basis matrices are jointly learned by introducing an autoregressive(AR)model.An experiment on the commonly used Cambridge hand-gesture database,which contains 900 videos in total separated into 9 different hand-gesture classes,with 100 videos in each.The videos in each gesture class were collected from 2 subjects and include videos captured under 5 different illumination conditions and containing 10 different arbitrary motions.Experimental result demonstrates that our proposed method reaches convergence within a small number of iterations during the training stage and achieves promising results compared with state-of-the-art methods.