Image processing methods, apparatus, devices, storage media and chips

By using image enlargement and super-resolution processing, the problem of subject position deviation was solved, improving user experience and image processing efficiency while reducing equipment requirements and costs.

CN122090427APending Publication Date: 2026-05-26BEIJING XIAOMI MOBILE SOFTWARE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING XIAOMI MOBILE SOFTWARE CO LTD
Filing Date
2024-11-26
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

When shooting videos or images, the position of the subject deviates significantly from the user's desired position, failing to meet requirements and making it impossible to adjust during post-editing, resulting in a poor user experience.

Method used

Image processing methods, including image expansion and super-resolution processing, are used to process the position information of the target object in the image frame into preset position information. A diffusion generation model is used for image filling and background matching, and the Unet network is used for noise reduction.

Benefits of technology

It enhances the user experience, improves the convenience and efficiency of image processing, reduces equipment requirements and costs, and is suitable for widespread application.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122090427A_ABST
    Figure CN122090427A_ABST
Patent Text Reader

Abstract

This disclosure relates to an image processing method, apparatus, device, storage medium, and chip. The method includes: acquiring a frame to be processed from a target video, the frame including a target object; performing image processing based on the target object in the frame to obtain a processed image frame; wherein the image processing includes at least image expansion processing, the image expansion processing being used to instruct the position information occupied by the target object in the image frame to be processed into preset position information. This solves the technical problems in related technologies, such as high equipment requirements and difficulty in widespread application.
Need to check novelty before this filing date? Find Prior Art