Human body behavior recognition method based on deep space-time inference network and electronic equipment

Abstract

The invention discloses a human body behavior recognition method based on a deep space-time inference network and electronic equipment, and the method comprises the steps of inputting target video data into a pre-trained deep space-time inference network model, and determining a first activation frequency corresponding to a clustering center of each node in the deep space-time inference network model; constructing a first behavior feature tree according to the first activation times; obtaining a first frequent sub-tree set from the first behavior feature tree by using a frequent sub-tree mining algorithm; and according to the first frequent sub-tree set, identifying human body behaviors in the target video data. Behavior recognition is performed through the human body dynamic behavior space-time features extracted by the deep space-time inference network and the different-level feature relationship reflected by the frequent subtree, the features of the human body parts in the behaviorsare concerned, and the association and coordination between the parts are concerned, so that the feature utilization rate in the network is improved, the recognition rate of the network is improved,and the human body behaviors in the video data can be quickly and accurately recognized.

Description

technical field [0001] The invention relates to the technical field of behavior recognition, in particular to a human behavior recognition method and electronic equipment based on a deep spatio-temporal reasoning network. Background technique [0002] With the development of computer video surveillance technology, intelligent surveillance systems are widely used in public places such as traffic intersections, airports, shopping malls, parking lots and commercial office buildings. Among them, human behavior recognition has become one of the most challenging problems in the field of intelligent monitoring. Its goal is to enable computers to recognize human behavior based on human body information, especially motion information. [0003] According to different recognition methods, the current research methods on behavior recognition can be divided into two categories: traditional methods and deep learning methods. Traditional methods mainly include methods based on template ma...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above-mentioned defects of the prior art, the present invention provides a human behavior recognition method and electronic equipment based on a deep spatio-temporal inference network, aiming at solving the problem that the existing human behavior recognition method based on deep learning can only recognize still pictures and cannot The problem of human behavior recognition in video data

Images