Video Block Transform Selection Without Index Signaling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video encoding standards face inefficiencies in transform index signaling, limiting the benefits of additional transforms due to increased signaling requirements, which hampers compression efficiency.

Innovation Solution

The method and apparatus determine transform indices at the decoder side by examining L-shaped reference decoded pixels surrounding the current block, eliminating the need for explicit signaling and allowing a wider range of transforms, thereby improving coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If additional transforms are introduced to improve compression efficiency, then coding efficiency is improved, but signaling requirements increase

Engineering Contradiction:

Improvecompression efficiencyVSAvoidtransform index signaling overhead

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The decoder autonomously determines the transform index by analyzing L-shaped reference decoded pixels surrounding the current block, eliminating the need for explicit transform index signaling from the encoder. This self-service mechanism allows the system to select appropriate transforms based on local image characteristics without increasing bitstream overhead.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

L-shaped reference decoded pixels serve as an intermediary between the image content and transform selection. By examining these reference pixels, the decoder can infer the appropriate transform type (e.g., DCT or DST) without direct signaling, using the reference pixels as a mediator to convey transform selection information implicitly.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Quantity of substance

If transform index signaling is reduced, then bitrate is reduced, but transform selection accuracy may deteriorate

Engineering Contradiction:

ImprovebitrateVSAvoidtransform selection accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The system changes the parameter used for transform selection from explicit transform index values to implicit characteristics derived from L-shaped reference decoded pixels. By analyzing the statistical properties and patterns of these reference pixels, the decoder can accurately determine the appropriate transform without relying on transmitted index signals, thus reducing bitrate while maintaining selection accuracy.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The L-shaped reference decoded pixels are decoded and available before the current block processing. This preliminary availability of reference data allows the decoder to pre-determine the transform index based on already-decoded neighboring pixels, ensuring accurate transform selection is made before encoding the current block without requiring additional signaling.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12556698B2Methods and apparatuses for encoding/decoding a video

Publication Date: 2026.02.17 INTERDIGITAL CE PATENT HOLDINGS SAS

AI summary

A method comprising decoding a video is provided, wherein decoding the video includes determining at least one transform for a block of the video based on at least one part of reconstructed pixels surrounding the block, decoding the block by applying the at least one determined transform. An apparatus for decoding the video is also provided. Corresponding method and apparatus for encoding a video are also provided.