Intra Prediction Mode Coding With a Third Most Probable Mode

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing HEVC standard's intra mode coding method is inefficient in terms of coding efficiency due to the limited use of most probable modes (MPMs) for predicting block modes, leading to increased data transmission requirements.

Innovation Solution

The method introduces a third most probable mode (MPM2) derived from the prediction modes of neighboring blocks, enhancing the likelihood that the current block's mode matches one of the derived modes, thereby reducing the overall coding cost by using a more economical encoding process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If only two most probable modes (MPM0 and MPM1) are derived from neighboring blocks for intra prediction mode coding, then the encoding process remains simple, but the coding efficiency is limited due to lower probability of matching the current block's actual mode

Engineering Contradiction:
Improvecoding efficiencyVSAvoidencoding process complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The invention segments the prediction mode selection by dividing it into two stages: first deriving two MPMs from neighboring blocks for quick comparison, then deriving a third MPM2 from the relationship between MPM0 and MPM1. This segmentation allows the system to handle the most common cases efficiently while providing additional modes for less common scenarios, thereby improving coding efficiency without proportionally increasing complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention performs preliminary derivation of three most probable modes (MPM0, MPM1, and MPM2) before the actual mode comparison and selection. By pre-calculating these modes based on neighboring block relationships, the system prepares a optimized set of candidate modes that are most likely to match the current block's actual mode, reducing the need for encoding additional mode information and improving overall coding efficiency.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If the number of supported intra prediction modes is increased for different coding unit sizes, then the prediction accuracy improves, but the data transmission requirements increase

Engineering Contradiction:
Improveprediction accuracyVSAvoiddata transmission requirements
Core Design Contradiction:
Manufacturing precisionVSLoss of information

Solution Approach 1:

The invention changes the parameter of mode derivation by introducing MPM2 which is derived from the relationship between MPM0 and MPM1 rather than directly from neighboring blocks. This parameter change in the derivation method allows the system to capture additional prediction patterns without significantly increasing the total number of modes, thereby improving prediction accuracy while controlling data transmission requirements.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The invention uses MPM2 as an intermediary mode that bridges the gap between the two directly derived MPMs and the actual current block mode. MPM2 is derived from the relationship between MPM0 and MPM1, serving as a mediator that captures additional prediction patterns. This intermediary approach improves prediction accuracy by providing an additional likely candidate without requiring a full increase in the number of supported modes.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4224857B1Deriving reference mode values and encoding and decoding information representing prediction modes
Publication Date: 2025.12.24 CANON KK
  • EP4224857B1 patent drawingFigure 1A~1B
  • EP4224857B1 patent drawingFigure 2
  • EP4224857B1 patent drawingFigure 3

AI summary

Reference prediction mode values, also referred to as most probable modes or MPMs, usable for encoding or decoding of a prediction mode related to a current coding unit, are derived. By deriving three MPMs instead of two for comparison with the prediction mode of the current coding block the coding efficiency is improved. This is due to the increase in the probability that the prediction mode of the current coding block corresponds to one of the derived most probable modes.