Identity authentication and basketball shooting action quality evaluation method and system fusing contour map and human body analysis map

By integrating contour maps and human anatomy diagrams, the problem of simultaneously achieving identity authentication and motion quality assessment in basketball shooting training was solved, improving the accuracy of identity recognition and the stability of motion assessment, and adapting to the differences among individuals.

CN122392123APending Publication Date: 2026-07-14HENAN UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HENAN UNIVERSITY
Filing Date
2026-04-07
Publication Date
2026-07-14

AI Technical Summary

Technical Problem

Existing technologies cannot simultaneously achieve high-accuracy identity authentication and objective and reliable shooting motion quality assessment in basketball free throw line and three-point line shooting training. They suffer from drawbacks such as poor dynamic adaptability of static facial recognition, insufficient representation of single modal features, and complex and poor adaptability of traditional joint angle calculation processes.

Method used

We employ a method that integrates contour maps and human body analysis maps. We use an improved Mask R-CNN and a fine-tuned CDGNet model for human detection and semantic segmentation. By combining feature extraction networks, feature fusion networks, and feature separation networks, we achieve collaborative modeling and evaluation of identity features and action features.

Benefits of technology

In non-confrontational fixed-point shooting scenarios, identity recognition and motion quality assessment are simultaneously achieved, improving the accuracy of identity recognition and the stability and adaptability of motion assessment, and adapting to individual differences in height, body type and shooting style.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122392123A_ABST
    Figure CN122392123A_ABST
Patent Text Reader

Abstract

This invention provides a method and system for identity authentication and basketball shooting motion quality assessment by fusing contour maps and human body analytical maps. The method includes: acquiring dynamic video of a basketball player shooting; extracting a sequence of human body images frame-by-frame from the video; performing human detection on the human body image sequence to obtain a sequence of human body foreground mask images; extracting human body contours based on the foreground mask images to obtain a sequence of human body contour images; performing semantic segmentation on the human body image sequence to obtain a sequence of human body analytical maps; inputting the human body contour image sequence and the human body analytical map sequence into a feature extraction network to obtain human body contour features and human body analytical features; jointly modeling the human body foreground mask image sequence, human body contour features, and human body analytical features through a feature fusion network to obtain cross-modal fusion features; and separating the cross-modal fusion features into identity features and motion features through a feature separation network to achieve athlete identity authentication and motion quality assessment.
Need to check novelty before this filing date? Find Prior Art