A Survey on Human Action Recognition based on Attention Mechanism

Human action recognition has always been an important challenge in the field of computer vision. Human action recognition has a good development prospect for technology applications in many fields such as artificial intelligence and behavior analysis. Based on the collected literature, this article first summarizes the data sets commonly used in the field of human action recognition, and then discusses the common methods of human action recognition based on traditional features and deep learning. Secondly, this paper reviews the principles and concepts of attention mechanism, and the types and applications of attention mechanism. Then this paper reviews the human action recognition methods based on the attention mechanism optimized in recent years and its comparison with other traditional methods on the data sets. Finally, the prospect and summary of the research on this subject are drawn.