Candidate Intra Prediction Modes for Shorter Video Codewords

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing demand for high resolution and high quality video data leads to increased data volume, resulting in higher costs for transmission and storage, necessitating more efficient video compression technologies.

Innovation Solution

A method and apparatus for encoding and decoding intra prediction modes using multiple candidate intra prediction modes, including determining and deriving candidate modes from neighboring blocks and prioritizing based on planar, DC, and vertical modes, to improve video encoding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If conventional intra prediction encoding is used, then video quality is maintained, but data volume and transmission/storage costs increase

Engineering Contradiction:

Improvetransmission and storage costsVSAvoidvideo compression efficiency

Core Design Contradiction:

Loss of energyVSManufacturing precision

Solution Approach 1:

The patent applies preliminary action by deriving candidate intra prediction modes from neighboring blocks before encoding the current block. The encoder prepares a list of candidate modes (including planar, DC, and directional modes) from adjacent blocks, then selects the most appropriate candidate for the current block. This pre-computation of candidate modes reduces the search space and enables more efficient encoding, directly addressing the contradiction by improving compression efficiency before the actual encoding process occurs.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If more candidate intra prediction modes are used, then encoding efficiency improves, but computational complexity increases

Engineering Contradiction:

Improveencoding efficiencyVSAvoidcomputational complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the intra prediction mode selection process into distinct stages: (1) deriving candidate modes from neighboring blocks, (2) selecting the best candidate based on rate-distortion optimization, and (3) encoding the selected mode. This segmentation allows the system to manage computational complexity by processing modes in discrete steps rather than evaluating all possible modes simultaneously, thereby improving encoding efficiency while controlling device complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by deriving only the necessary candidate modes from neighboring blocks rather than evaluating all possible intra prediction modes. The encoder selectively derives planar, DC, and directional modes from adjacent blocks based on their availability and relevance, performing partial computation that is sufficient to achieve good encoding efficiency without the full computational burden of evaluating every possible mode.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12501055B2Methods and apparatuses of encoding/decoding intra prediction mode using candidate intra prediction modes

Publication Date: 2025.12.16 DOLBY LABORATORIES LICENSING CORP

AI summary

A method and an apparatus of encoding/decoding intra prediction mode using a plurality of candidate intra prediction modes are disclosed. The method includes deriving three candidate intra prediction modes about a current block and deriving an intra prediction mode of the current block.