Intra Prediction Mode List Generation for Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding techniques face challenges in reducing complexity and improving coding efficiency, particularly in generating a most probable mode (MPM) list for intra prediction, which is essential for efficient video compression as video data volumes increase with higher resolutions and frame rates.
Innovation Solution
A method and apparatus that derive an intra prediction mode for a current block based on neighboring blocks, generate an MPM list using a histogram of mode combinations, and classify intra prediction modes into groups to enhance video encoding and decoding efficiency, allowing for the generation of an MPM list without relying on neighboring block information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MPM list is generated using traditional methods based on neighboring block information, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing histogram data of intra prediction mode combinations during an offline training phase. During actual video encoding/decoding, the system retrieves pre-computed histogram information to generate MPM lists, eliminating the need for real-time complex calculations based on neighboring block configurations. This significantly reduces decoding complexity while maintaining coding efficiency.
Solution Approach 2:
The patent uses copying by creating a simplified representation of the complex MPM list generation process through histogram-based probability distributions. Instead of directly processing complex neighboring block relationships in real-time, the system copies the essential statistical patterns into pre-computed histogram data structures, which can be quickly queried and applied during encoding/decoding operations.
2Measurement precision
If video resolution and frame rate are increased, then video quality is improved, but data volume increases requiring more storage and processing resources
Solution Approach 1:
The patent extracts the essential information needed for prediction from the complex video data by using histogram-based statistical models that capture the probability distributions of intra prediction modes. This extraction allows the system to maintain high video quality through accurate prediction while reducing the amount of data that needs to be processed and stored, as only the histogram parameters and MPM list indices are transmitted rather than full prediction data.
Solution Approach 2:
The patent applies parameter changes by transforming the representation of prediction information from detailed neighboring block configurations to simplified histogram probability parameters. This parameter transformation enables efficient compression of video data at high resolutions and frame rates by reducing the information density requirements while preserving the essential statistical characteristics needed for accurate intra prediction.
Data Source
AI summary
A video encoding/decoding method and a video encoding/decoding device are provided. The video decoding method according to the present disclosure includes determining at least two intra prediction modes based on intra prediction modes of neighboring blocks adjacent to a current block. The video decoding method also includes deriving a most probable mode (MPM) list based on the at least two intra prediction modes. The video decoding method also includes deriving an intra prediction mode of the current block based on the MPM list. The video decoding method also includes generating a prediction block of the current block based on the intra prediction mode. Candidate modes in the MPM list can be configured using an intra prediction mode list generated based on the at least two intra prediction modes.