Pooling the Convolutional Layers in Deep ConvNets for Action Recognition
This work addresses action recognition in videos, an incremental improvement over existing methods by better utilizing temporal information.
The paper tackled the problem of deep video representation for action recognition by proposing a framework that pools convolutional layers from spatial and temporal ConvNets using temporal pooling strategies and VLAD encoding, achieving state-of-the-art accuracy of 93.78% on UCF101 and competitive 65.62% on HMDB51.
Deep ConvNets have shown its good performance in image classification tasks. However it still remains as a problem in deep video representation for action recognition. The problem comes from two aspects: on one hand, current video ConvNets are relatively shallow compared with image ConvNets, which limits its capability of capturing the complex video action information; on the other hand, temporal information of videos is not properly utilized to pool and encode the video sequences. Towards these issues, in this paper, we utilize two state-of-the-art ConvNets, i.e., the very deep spatial net (VGGNet) and the temporal net from Two-Stream ConvNets, for action representation. The convolutional layers and the proposed new layer, called frame-diff layer, are extracted and pooled with two temporal pooling strategy: Trajectory pooling and line pooling. The pooled local descriptors are then encoded with VLAD to form the video representations. In order to verify the effectiveness of the proposed framework, we conduct experiments on UCF101 and HMDB51 datasets. It achieves the accuracy of 93.78\% on UCF101 which is the state-of-the-art and the accuracy of 65.62\% on HMDB51 which is comparable to the state-of-the-art.