Intra Prediction MPM List Construction for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image/video compression techniques are inefficient for high-resolution, high-quality images/videos, leading to increased transmission and storage costs, and fail to effectively compress and transmit diverse content types like VR/AR and hologram images/videos.
Innovation Solution
An image decoding method that derives a Most Probable Mode (MPM) list for a current block by considering neighboring blocks, constructing an MPM list based on candidate intra prediction modes from left and upper neighboring blocks to improve coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional compression techniques are used for high-resolution images/videos, then transmission and storage costs increase, but compression efficiency remains insufficient
Solution Approach 1:
The patent changes the parameter of intra prediction mode selection by introducing an MPM list that adapts to different block characteristics. Instead of using a fixed prediction mode, the system dynamically selects from multiple candidate modes based on neighboring block information, improving compression efficiency for high-resolution content without proportionally increasing transmission costs
Solution Approach 2:
The patent implements dynamic adaptation in the intra prediction process by constructing MPM lists that vary according to the specific characteristics of neighboring blocks. This dynamic approach allows the system to optimize prediction accuracy for each block individually, enhancing overall compression efficiency for diverse high-resolution content types including VR/AR and hologram images
2Measurement precision
If more intra prediction modes are considered for accurate representation, then coding efficiency improves, but computational complexity increases
Solution Approach 1:
The patent segments the set of all possible intra prediction modes into two categories: MPM (Most Probable Mode) candidates and non-MPM candidates. By dividing the prediction mode selection process into these segments, the system can focus computational resources on evaluating a limited set of most likely modes first, achieving accurate prediction without the need to exhaustively search all possible modes
Solution Approach 2:
The patent performs preliminary action by pre-constructing the MPM list based on neighboring block information before the actual prediction mode selection. This preliminary construction of candidate modes based on spatial correlation with adjacent blocks allows the decoder to quickly identify the most probable prediction mode without performing complex computations, thereby improving accuracy while controlling computational complexity
3Productivity
If MPM list is constructed using only traditional neighboring blocks, then decoding simplicity is maintained, but coding efficiency for diverse content types is insufficient
Solution Approach 1:
The patent applies local quality by constructing different MPM lists based on the specific characteristics and availability of different neighboring blocks. Instead of using a uniform approach for all blocks, the system adapts the MPM list construction to local conditions, selecting appropriate neighboring blocks (such as left, upper, or other spatial neighbors) based on their availability and relevance to the current block being decoded
Solution Approach 2:
The patent achieves universality by designing an MPM list construction mechanism that can handle multiple content types (conventional images, VR/AR content, hologram images) through a unified framework. The same basic MPM list construction process adapts to different content characteristics by selecting appropriate neighboring blocks, providing improved coding efficiency across diverse content types while maintaining decoding simplicity through a consistent algorithmic approach
Data Source
AI summary
An image decoding method according to the present disclosure comprises deriving a first candidate intra prediction mode based on a first neighboring block of a current block; deriving a second candidate intra prediction mode based on a second neighboring block of the current block; constructing MPM (Most Probable Mode) list of the current block based on the first candidate intra prediction mode and the second candidate intra prediction mode; deriving an intra prediction mode for the current block based on the MPM list; and generating a prediction sample for the current block based on the intra prediction mode, wherein the first neighboring block is a left neighboring block located at the lowermost side among neighboring blocks adjacent to a left boundary of the current block, and wherein the second neighboring block is an upper neighboring block located at the rightmost side among neighboring blocks adjacent to an upper boundary of the current block.