Robot gait learning method and system based on video stream and storage medium

The invention discloses a robot gait learning method and system based on video streaming and a storage medium, and the method comprises the steps: firstly, respectively collecting the motion video streaming of a quadruped organism moving in different environments, responding to a key frame image, extracting key pixel points of the skeleton of the quadruped organism in the key frame image, using a neural network for carrying out deep learning on different gait information corresponding to different environments, recognizing environment information of a current frame image in visual equipment, establishing a discretization gait model according to foot point coordinates in the gait information, and solving foot point coordinates of a next foot point when the four-foot bionic robot moves in the discretization gait model. According to the method, feature points are extracted through massive videos to obtain skeleton actions, so that stable bionic gait machine learning is carried out by utilizing video feature poin