Intra Prediction Mode List Segmentation for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding technologies face inefficiencies in determining the most probable mode lists for intra prediction, which affects the coding efficiency of video data.
Innovation Solution
The techniques involve constructing a general most probable mode list containing N entries, with a planar mode as the first entry, and then creating primary and secondary lists by selecting intra prediction modes from neighboring blocks and offset modes, to determine the current intra prediction mode for encoding or decoding video data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional intra prediction mode determination methods are used, then the process is simpler, but coding efficiency is reduced
Solution Approach 1:
The patent divides the most probable mode list generation into two separate lists: primary MPM list and secondary MPM list. The primary list contains modes from direct neighboring blocks, while the secondary list contains offset modes derived from neighboring block modes. This segmentation allows the decoder to first check the simpler primary list before considering the more complex secondary list, improving coding efficiency without requiring the entire complex structure to be processed uniformly.
Solution Approach 2:
The patent performs preliminary action by constructing the primary MPM list first using direct neighboring block modes, then uses these results to generate the secondary MPM list with offset modes. This preliminary generation of the primary list provides a foundation that improves overall coding efficiency, as many cases can be resolved with the simpler primary list before needing to process the more complex secondary list.
2Productivity
If a comprehensive most probable mode list is constructed, then coding efficiency improves, but processing time increases
Solution Approach 1:
The patent segments the comprehensive MPM list into primary and secondary lists with different priorities. The decoder processes the primary list first (faster, simpler modes from neighboring blocks), and only if needed proceeds to the secondary list (more complex offset modes). This segmentation reduces average processing time while maintaining coding efficiency, as not all modes need to be fully processed in every case.
Solution Approach 2:
The patent implements partial action by allowing the decoder to stop processing after finding a match in the primary MPM list, without needing to generate or process the entire secondary list. This partial processing approach achieves most of the coding efficiency benefit while avoiding the full time cost of generating comprehensive mode lists in all cases.
Data Source
AI summary
A video coder may code a block of video data using an intra prediction mode determined from a most probable mode list. The video coder may construct a general most probable mode list containing N entries, wherein the N entries of the general most probable mode list are intra prediction modes, and wherein a planar mode is an ordinal first entry in the general most probable mode list, construct a primary most probable mode list from a first Np entries in the general most probable mode list, where Np is less than N, and construct a secondary most probable mode list from a remaining (N−Np) entries in the general most probable mode list. The video coder may then determine a current intra prediction mode for a current block of video data using the primary most probable mode list or the secondary most probable mode list.