Motion scoring method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a motion scoring method and device, electronic equipment and a readable storage medium. The method comprises the following steps: determining a target search range corresponding to the tth frame of reference video image in a reference video image group, according to the target search range, determining a plurality of frames of simulated video images from asimulated video image group, determining a target imitation video image from the multiple frames of simulated video images, wherein the similarity between the key points of the target simulation videoimage and the key points of the tth frame of reference video image is the maximum similarity between the similarity between the key points of each frame of simulation video image and the key points of the tth frame of reference video image; determining a time difference between a time point corresponding to the target simulation video image and a time point corresponding to the tth frame of reference video image, and taking the time difference as the time difference corresponding to the tth frame of reference video image, and determining a scoring result corresponding to the simulated video image group according to the time difference corresponding to each frame of reference video image, thereby improving the precision of the scoring result.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to an action scoring method, device, electronic equipment and readable storage medium. Background technique [0002] With the improvement of computer computing power and the development of deep learning field, the key point recognition algorithm based on deep learning has been proposed. The network model is trained through a large amount of labeled data, so that the network model can recognize key points of the human body in various poses. Identifying key points of the human body can be applied to, for example, when a user imitates a certain dance, and evaluates the user according to the similarity of the user's imitation. [0003] At present, by obtaining the imitation image of the user, key points are determined from the imitation image, such as points at joints such as shoulders, knees, and elbows, and the coordinates of each key point are obtained as the imitation coordinates...

AI Technical Summary

Problems solved by technology

However, the accuracy of scoring user imitation actions based on similarity is not high enough, so how to provide a higher-precision scoring scheme needs to be solved urgently

Images