Image block prediction method, device and electronic equipment
By combining reference row determination methods with multiple intra-frame prediction modes, the accuracy of image patch prediction is improved, solving the problem of inaccurate prediction in existing technologies.
CN119728988BActive Publication Date: 2026-07-24VIVO MOBILE COMM CO LTD
View PDF 1 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- VIVO MOBILE COMM CO LTD
- Filing Date
- 2023-09-28
- Publication Date
- 2026-07-24
AI Technical Summary
Technical Problem
In existing intra-prediction mode fusion techniques, determining the position of the reference line solely through a single intra-prediction mode leads to inaccurate predictions.
Method used
By obtaining the first reference row corresponding to the first prediction mode in at least two intra-frame prediction modes of the target image block, and combining it with at least one second prediction mode to determine the second reference row corresponding to the second prediction mode, the prediction block is fused.
Benefits of technology
It improves the prediction accuracy of multiple intra-frame prediction mode fusion techniques and avoids the error when using only the second prediction mode to determine the reference line.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN119728988B_ABST
Abstract
The application discloses an image block prediction method and device and electronic equipment, and belongs to the technical field of coding and decoding. The method of the application embodiment comprises the following steps: acquiring a first reference line corresponding to a first prediction mode in at least two intra-prediction modes corresponding to a target image block; determining a second reference line corresponding to at least one second prediction mode according to the first prediction mode and the at least one second prediction mode, wherein the second prediction mode is one prediction mode in the at least two intra-prediction modes except the first prediction mode; and fusing a first prediction block and at least one second prediction block to determine a target prediction block corresponding to the target image block, wherein the first prediction block is determined based on the first prediction mode and the first reference line, and each second prediction block is determined based on one second prediction mode and the second reference line.
Need to check novelty before this filing date? Find Prior Art