Character animation generation method and device, electronic equipment and storage medium

By acquiring the identity information of the target character and generating image frames of a 2D video animation using the target model, the uniformity of the character's facial features is maintained by using Identity Locking Vector (ICM), which solves the problem of facial distortion and improves the efficiency and consistency of animation generation.

CN122176137APending Publication Date: 2026-06-09GUANGZHOU BOGUAN TELECOMM TECH LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
GUANGZHOU BOGUAN TELECOMM TECH LTD
Filing Date
2026-01-27
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies often result in severe facial distortion of characters when generating 2D video animations, leading to unstable appearance features and affecting the efficiency and consistency of animation generation.

Method used

By acquiring the identity indication information of the target character, the target model generates image frames of the character animation based on the identity indication information, the first cue word, and the second cue word, and generates the animation based on the image frames. Identity Locking Vector (ICM) is used to maintain the uniformity of the character's facial features and the stability of its appearance features.

Benefits of technology

It improves the consistency of characters in animation, ensures the stability of character appearance features, and improves the efficiency of animation generation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122176137A_ABST
    Figure CN122176137A_ABST
Patent Text Reader

Abstract

The application discloses a role animation generation method and device, electronic equipment and a storage medium, comprising: obtaining identity indication information of a target role, the identity indication information comprising at least one role identity feature of the target role, and one role identity feature indicating one visual performance effect of the target role; obtaining a role expression reference image matched between the target role and the identity indication information; outputting at least one image frame of the target role based on the identity indication information, a first prompt word and a second prompt word through a target model; and generating a role animation of the target role based on each image frame; the application can accurately maintain the unity of the facial features of the role and the stability of the appearance features of the role in the continuous image frame sequence generation process of the animation, thereby improving the role consistency of the role in the animation and effectively improving the generation efficiency of the animation.
Need to check novelty before this filing date? Find Prior Art