基于注意力机制与三维卷积聚合的动作识别

Action Recognition Based on Attention Mechanism and Three-dimensional Convolutional Aggregation

由于视频数据大量,视觉内容丰富,如何有效地提取视频中的时间特征,有效地融合时空特征是动作识别中的一个难题。针对这些困难,提出了一种基于注意力机制和三维卷积聚合的动作识别新算法。为了验证该算法的有效性,在大型公共行为数据集UCF101上进行了验证。实验结果表明,该算法具有良好的时间特征建模能力,有效地提高了动作识别的精度。

Due to the large amount of video data and rich visual content,how to effectively extract the temporal features in the video and effectively fuse the spatiotemporal features is a difficult problem in action recognition.To solve these difficulties,a new algorithm for action recognition based on attention mechanism and three-dimensional convolutional aggregation is proposed.To verify the effectiveness of the algorithm,it is verified on the large public behavior dataset UCF101.The experimental results show that the algorithm has good temporal feature modeling ability,which effectively improves the accuracy of action recognition.