Wide-Angle Intra Prediction Mapping for Non-Square Coding Blocks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding technologies face inefficiencies in intra prediction due to the limited number of directions and bit representation of less likely prediction modes, leading to suboptimal compression ratios and increased bit usage.

Innovation Solution

The proposed method involves remapping intra prediction modes for non-square block partitions using wide-angle modes based on aspect ratios, with a predefined number of mode replacements and angle parameters determined by a lookup table, allowing for more efficient encoding and decoding of video data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional intra prediction modes with limited directions are used, then device complexity is reduced, but compression ratio deteriorates due to insufficient prediction accuracy

Engineering Contradiction:
Improvecompression ratioVSAvoidnumber of prediction modes
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies dynamics by making the intra prediction mode selection adaptive rather than static. The encoder dynamically chooses between planar/DC modes and wide-angle modes based on the block's aspect ratio, allowing the prediction system to adapt to different block shapes and optimize compression performance accordingly.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of prediction direction angles by introducing wide-angle modes with angles outside the traditional ±45° range. This parameter expansion allows better modeling of diagonal edges in non-square blocks, improving prediction accuracy and compression ratio without requiring a complete redesign of the intra prediction framework.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If more intra prediction modes with extended angles are added, then prediction accuracy improves, but bit usage increases for mode representation

Engineering Contradiction:
Improveprediction accuracyVSAvoidbit usage for mode encoding
Core Design Contradiction:
Manufacturing precisionVSLoss of information

Solution Approach 1:

The patent applies local quality by applying wide-angle prediction modes selectively only to non-square blocks where they provide benefit, rather than universally. Square blocks continue to use traditional planar/DC modes, avoiding unnecessary complexity and bit overhead for modes that would not improve prediction accuracy in that context.

Inventive Principle:
Principle #3Local quality

3Productivity

If intra prediction is applied to non-square blocks with traditional modes, then coding simplicity is maintained, but coding efficiency deteriorates due to mismatched prediction directions

Engineering Contradiction:
Improvecoding efficiencyVSAvoidmode remapping logic
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining the mapping between traditional intra prediction modes and wide-angle modes based on aspect ratio thresholds. This lookup table approach allows the encoder to quickly determine the appropriate mode without complex real-time calculations, reducing computational overhead while improving coding efficiency.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20260075244A1Mapping wide angle modes to intra prediction modes for a partition of a coding block
Publication Date: 2026.03.12 TENCENT AMERICA LLC
  • US20260075244A1 patent drawing
  • US20260075244A1 patent drawing
  • US20260075244A1 patent drawing

AI summary

Aspects of the disclosure provide methods and apparatuses for video encoding/decoding. In some examples, an apparatus for video decoding includes receiving circuitry and processing circuitry. The processing circuitry decodes prediction information of a current block from a coded video bitstream. In some embodiments, the processing circuitry determines a first sub-partition and a second sub-partition of the current block based on the decoded prediction information, and then reconstructs the first sub-partition and the second sub-partition of the current block based on at least a neighboring sample of the current block that is out of a neighboring area of at least one of the first sub-partition and the second sub-partition.