Intra Mode Propagation via Universal Map for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently determining intra prediction modes for video blocks, particularly in accurately representing likely and unlikely prediction directions, which affects compression efficiency and redundancy reduction.
Innovation Solution
The method involves determining an intra prediction mode for a current block by searching candidate matching areas in a reconstructed picture, calculating cost values between template regions, and selecting the area with the minimum cost value, using a universal intra mode map to divide areas into sub-regions associated with specific modes, and propagating intra modes from neighboring blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional intra coding without intra prediction is used, then the coding process is simple, but the compression ratio is poor and redundancy is not effectively reduced
Solution Approach 1:
The patent applies preliminary action by pre-defining a universal intra mode map that divides the picture into multiple regions, each associated with specific intra prediction modes. This pre-organization allows the decoder to quickly determine prediction modes without complex calculations during decoding, thereby improving compression efficiency while maintaining simple decoding processes.
Solution Approach 2:
The patent implements local quality by assigning different intra prediction modes to different spatial regions of the picture based on the universal intra mode map. Each region is optimized with prediction modes suitable for its local characteristics, allowing more accurate prediction in areas with specific directional patterns while keeping other areas simpler, thus improving overall compression ratio.
2Productivity
If multiple intra prediction modes are used to accurately represent different directions, then the compression efficiency improves, but the device complexity increases
Solution Approach 1:
The patent reduces device complexity through preliminary action by pre-establishing the universal intra mode map structure with defined regions and their associated prediction modes. This pre-configuration eliminates the need for complex real-time analysis during decoding, as the decoder simply needs to look up the appropriate mode from the pre-defined map based on the current block's position.
Solution Approach 2:
The patent applies segmentation by dividing the picture into multiple regions through the universal intra mode map, where each region is associated with specific intra prediction modes. This segmentation allows the system to use different prediction strategies for different areas, improving compression efficiency while keeping the complexity manageable through structured regional division rather than analyzing every block individually.
3Adaptability or versatility
If intra prediction modes are determined without a universal map structure, then the decoding process is flexible, but the accuracy of prediction mode selection is reduced
Solution Approach 1:
The patent achieves universality by creating a universal intra mode map that can be applied across different picture types and block sizes. This single unified structure serves multiple functions: it provides region-based mode assignment, maintains decoding flexibility through standardized procedures, and improves accuracy by leveraging spatial correlations. The universal map works consistently across various scenarios without requiring separate complex determination logic for each case.
Data Source
AI summary
Coded information of (i) a current block of a picture and (ii) a reconstructed area of the picture is received. A matching area of the current block in the reconstructed area of the picture is determined. A first corresponding position of the current block is determined inside the current block. The first corresponding position includes a first coordinate value on a first axis and a second coordinate value on a second axis relative to a first reference point on the current block. A corresponding block of the current block is determined in the matching area. The corresponding block includes a second corresponding position that has the first coordinate value on the first axis and the second coordinate value on the second axis relative to a second reference point on the matching area. An intra prediction mode of the current block is determined based on the corresponding block.