Human body behavior recognition method and system based on self-attention enhancement graph neural network

The invention belongs to the technical field of human behavior recognition in the aspect of computer vision, and particularly relates to a human behavior recognition method and system based on a self-attention enhancement graph neural network. The method comprises the following steps: acquiring joint information representation, skeleton information representation and angle information representation in skeleton information; based on the joint information representation, the skeleton information representation and the angle information representation, using a graph convolution module to aggregate features of adjacent joints to obtain a first feature graph; based on the first feature map, extracting joint features and skeleton features of different scales by using a multi-scale enhanced image convolution block; based on the first feature map, a multi-head attention block is adopted, and fine-grained action angle features are extracted in the time dimension; on the basis of the joint features, the skeleton featu