Image processing method, apparatus, device, medium and product
By determining the fusion weights through optical flow deformation and the correlation of jitter image features, the artifact problem of the TAA method under complex image content is solved, and higher quality image processing is achieved.
CN122265056APending Publication Date: 2026-06-23MOORE THREADS TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- MOORE THREADS TECH CO LTD
- Filing Date
- 2026-05-26
- Publication Date
- 2026-06-23
AI Technical Summary
Technical Problem
Existing temporal antialiasing (TAA) methods suffer from artifacts due to biased fusion weights when dealing with complex image content, making it difficult to effectively remove aliasing.
Method used
By deforming the image based on optical flow information and determining the fusion weight by combining the feature correlation of the jittery image, the spatial alignment of the historical image and the current image is achieved, and artifacts are removed by weighted fusion.
Benefits of technology
It effectively removes artifacts, improves image quality, and enhances anti-aliasing, especially in complex image content where it can better smooth jagged edges.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122265056A_ABST
Abstract
This application discloses an image processing method, apparatus, device, medium, and product, relating to the field of image technology. The method includes: performing deformation processing on an (i-1)th output image based on optical flow information to obtain an (i-1)th deformed image, where the (i-1)th output image is the image rendered and displayed at time (i-1), and i is a positive integer; determining a fusion weight based on the feature correlation between the image features of the (i-1)th deformed image and the image features of the i-th jittered image, where the i-th jittered image is obtained by jitter sampling of the original image at time (i); and fusing the (i-1)th deformed image and the i-th jittered image based on the fusion weight to obtain the i-th output image.
Need to check novelty before this filing date? Find Prior Art