Video Coding Edge-Based Sub-Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The intra sub-partition (ISP) prediction process in video coding is computationally intensive and inefficient, as it requires calculating and comparing RD costs for all intra prediction modes for each sub-CU, leading to increased processing load and potential reduction in prediction accuracy due to shared intra prediction modes across sub-CUs.
Innovation Solution
A method that divides coding unit blocks based on edge direction and strength, allowing sub-coding unit blocks to reference each other within the same coding unit block, thereby reducing the need for extensive RD cost calculations and processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If the ISP prediction process is used to reduce code size by sharing intra prediction modes across sub-CUs, then code size is reduced, but processing load increases due to requiring RD cost calculations for all intra prediction modes
Solution Approach 1:
The patent performs preliminary analysis of edge directions and edge strengths in the coding unit block before determining sub-CU division and intra prediction modes. By pre-calculating edge characteristics and using them to guide the ISP prediction process, the method avoids performing full RD cost calculations for all intra prediction modes, thereby reducing processing load while maintaining coding efficiency
Solution Approach 2:
The patent applies different intra prediction modes to different sub-CUs based on local edge characteristics. By analyzing edge directions and strengths in specific regions and assigning appropriate prediction modes locally, the method achieves accurate prediction without requiring exhaustive RD cost calculations across all modes, thus balancing code size reduction with processing efficiency
2Measurement precision
If RD cost calculations are performed for all intra prediction modes to ensure optimal prediction accuracy, then prediction accuracy is improved, but processing time and complexity increase
Solution Approach 1:
The patent performs partial RD cost calculations only for candidate intra prediction modes that are selected based on preliminary edge analysis, rather than calculating RD costs for all 33 (HEVC) or 65 (VVC) intra prediction modes. This partial action approach maintains prediction accuracy for the most suitable modes while significantly reducing processing time
Solution Approach 2:
By performing preliminary edge direction and edge strength analysis before intra prediction mode selection, the patent pre-identifies candidate modes that are most likely to be optimal. This preliminary action eliminates the need for exhaustive RD cost calculations, reducing processing time while preserving prediction accuracy for the selected candidate modes
Data Source
AI summary
A coding method encodes an image by dividing the image into blocks, and comprises: dividing a coding target image into a plurality of coding unit blocks; determining whether or not to encode by dividing the coding unit block into a plurality of sub-coding unit blocks, on a basis of an edge direction and an edge strength of an edge obtained for each pixel in the coding unit block; and encoding, in a case of determining to divide the coding unit block into a plurality of sub-coding unit blocks in the determining step, a first sub-coding unit block by referencing a second sub-coding unit block inside the same coding unit block as the first sub-coding unit block.