Transform Pair Selection for Non-Redundant Video Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video encoding and decoding technologies face challenges in achieving high compression efficiency due to redundancy in transform set selection, leading to inefficiencies in video coding schemes.

Innovation Solution

A method and apparatus that involve obtaining alternative transform sets with non-redundant transform pairs, allowing for selection of optimal transform pairs for encoding and decoding processes, and signaling these selections through indices in the bitstream.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple transform sets are provided for selection, then compression efficiency is improved, but redundancy in transform set selection increases

Engineering Contradiction:
Improvecompression efficiencyVSAvoidredundancy
Core Design Contradiction:
ProductivityVSLoss of substance

Solution Approach 1:

The patent extracts and removes redundant transform pairs from the alternative transform sets, keeping only non-redundant transform pairs. This is achieved by comparing transform pairs across different transform sets and eliminating duplicates, thereby reducing redundancy while preserving the beneficial diversity of transform options for improved compression efficiency

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent merges multiple transform sets by combining their transform pairs into a unified structure, then applies deduplication. This merging process allows the system to leverage the strengths of multiple transform sets while eliminating redundancy through the non-redundant transformation, achieving both compression efficiency and reduced redundancy

Inventive Principle:
Principle #5Merging (Combining)

2Adaptability or versatility

If transform pairs are added to alternative transform sets, then transform selection flexibility is improved, but complexity of transform set management increases

Engineering Contradiction:
Improvetransform selection flexibilityVSAvoidtransform set management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent extracts only the essential non-redundant transform pairs from multiple transform sets, removing duplicate entries. This extraction process maintains transform selection flexibility by preserving diverse transform options while reducing the overall size and management complexity of the transform sets

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter of transform set composition by transforming it from a collection that may contain duplicates to a streamlined collection of non-redundant transform pairs. This parameter change optimizes the balance between flexibility and complexity by reorganizing the transform set structure

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4661395A1Encoding and decoding methods using multiple transform set selection and corresponding apparatuses
Publication Date: 2025.12.10 INTERDIGITAL CE PATENT HOLDINGS SAS
  • EP4661395A1 patent drawingFigure 1
  • EP4661395A1 patent drawingFigure 2
  • EP4661395A1 patent drawingFigure 3

AI summary

A decoding method is disclosed. An initial transform set is obtained. One or more alternative transform sets are obtained wherein each alternative transform set comprising one or more transform pairs. Transform pairs of the one or more alternative transform sets are added into a final alternative transform set only if non-redundant with a transform pair of the initial transform set. A transform pair is selected from any one of the initial transform set and the final alternative transform set. An image block is reconstructed using the selected transform pair.