Video detection and rendering method, human motion detection method and electronic device

By smoothing the target human key point information and optimizing the model in the mobile video detection method, the problem of key point jitter in mobile videos is solved, and the stability of key points between video frames is improved.

CN114998211BActive Publication Date: 2025-10-03ALIBABA DAMO (HANGZHOU) TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202210465005.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-29
Publication Date
2025-10-03
Estimated Expiration
2042-04-29

AI Technical Summary

Technical Problem

The existing mobile terminal key point detection method suffers from severe key point jitter when detecting key points of the target human body in the video, and the key point stability between video frames is poor.

Method used

By obtaining the target human key point information of the video frame image in the video to be detected and performing smoothing processing, the target human key point image is obtained using the smoothed information, and the target human key point detection model is optimized through training data to improve the key point stability between video frames.

Benefits of technology

The jitter phenomenon of target human key point detection in mobile videos is reduced, and the key point stability between video frames is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114998211B_ABST
    Figure CN114998211B_ABST
Patent Text Reader

Abstract

The present application provides a video detection and rendering method, a human motion detection method, a detection model acquisition method, and an electronic device, wherein the video detection method comprises: obtaining target human key point information corresponding to the target human body in the video frame image in the video to be detected; smoothing the obtained target human key point information corresponding to the target human body in the video frame image in the video to be detected; and obtaining the target human key point image corresponding to the video frame image in the video to be detected based on the target human key point information corresponding to the target human body in the video frame image in the video to be detected after smoothing. This video detection method reduces the key point jitter phenomenon that occurs when the key point detection model performs key point detection on the target human body in the video on the mobile terminal by smoothing the target human key point information, thereby improving the key point stability between video frames in the video.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Key point position determination method and device, electronic equipment and storage medium

    CN111126101A

  • Method and a system for realizing posture evaluation guidance of a sports coach.

    CN112784699A

  • Human body key point detection model training method and device, detection method and device

    CN113361380A

  • Behavior prediction method and device based on depth vision, apparatus and storage medium

    CN113705534A