Action teaching error correction system and method based on fine-grained timing alignment

By combining fine-grained temporal alignment and dynamic temporal warping algorithms with a parameter library of school age groups and sports projects and error causal association rules, the accuracy and real-time performance of motion correction in AI sports teaching are achieved. This solves the problems of coarse temporal alignment granularity and low computational efficiency in existing technologies, and adapts to the teaching needs of different school age groups and sports projects.

CN122076005BActive Publication Date: 2026-07-24RONGMENGYUESHI (SHANGHAI) SPORTS TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
RONGMENGYUESHI (SHANGHAI) SPORTS TECHNOLOGY CO LTD
Filing Date
2026-04-22
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing AI-based sports teaching methods for motion correction have coarse timing alignment granularity, cannot accurately locate the timing errors, do not adapt to the differences in timing characteristics of different age groups and sports, lack error correlation analysis between timing stages, have low computational efficiency, are difficult to achieve real-time correction, and the correction results lack subject authority.

Method used

The motion teaching error correction system based on fine-grained temporal alignment breaks down standard sports movements into multiple fine-grained temporal stages, sets temporal time windows and posture key point thresholds, constructs a two-dimensional temporal alignment parameter library for school age groups and sports projects, and combines dynamic temporal regularization algorithms and error causal association rules to perform dynamic temporal alignment and error correction prompt generation. It adopts terminal lightweight computing and hybrid computing modes to ensure real-time performance.

Benefits of technology

It enables precise identification of movement errors and their root causes, is adaptable to multiple age groups and sports, improves the efficiency and real-time nature of error correction in sports movement teaching, reduces costs, and enhances the accuracy and practicality of error correction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122076005B_ABST
    Figure CN122076005B_ABST
Patent Text Reader

Abstract

The application discloses a motion teaching error correction system and method based on fine-grained timing alignment, and belongs to the technical field of artificial intelligence sports teaching, and comprises the following steps: standard sports motions are divided into multiple fine-grained timing stages, timing time windows, posture key point thresholds and error causal correlation rules between stages are set, and a school age-sports project two-dimensional timing alignment parameter library is constructed; timing frame data of student motions to be corrected are acquired, fine-grained timing alignment with standard motions is realized through a dynamic timing regularization algorithm; error points are detected based on posture key point thresholds, and error root stages are traced back in combination with correlation rules; lightweight feature dimension reduction and error correction prompt generation are performed on a terminal, real-time monitoring and calculation delay are performed, and when a threshold is exceeded, a hybrid calculation mode combining edge precalculation and cloud optimization is switched to, so that real-time performance is ensured; the application can accurately locate motion errors and roots, is suitable for multiple school ages and sports projects, and improves sports motion teaching error correction efficiency and real-time performance.
Need to check novelty before this filing date? Find Prior Art