Shape-Adaptive DCT for Geometric Blocks With Adaptive Region Partitioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression methods using rectangular block discrete cosine transform (B-DCT) are inefficient for non-rectangular blocks, requiring unnecessary computing resources and failing to represent pixel information effectively, leading to increased complexity and reduced compression efficiency.

Innovation Solution

Implementing shape adaptive discrete cosine transform (SA-DCT) for geometric partitioning with an adaptive number of regions, allowing selection between B-DCT and SA-DCT based on prediction error for each region, and signaling this choice in the bitstream for efficient encoding and decoding of non-rectangular blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a fixed number of regions is used for geometric partitioning, then the encoding process is simple and fast, but the transform efficiency and compression performance are limited

Engineering Contradiction:
Improveencoding speedVSAvoidtransform efficiency
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent implements a dynamic region partitioning system where the number of regions is not fixed but adaptively determined based on the characteristics of the input data block. The encoder evaluates various partitioning configurations and selects the optimal number of regions dynamically, allowing the system to balance between encoding complexity and transform efficiency for different types of data.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the parameter of region count from a static value to a dynamic variable that can take different values based on data characteristics. By adjusting the number of regions according to the specific input block properties, the transform efficiency is improved without requiring a complete redesign of the encoding framework.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If the number of regions is increased to improve transform efficiency, then the compression performance improves, but the encoding complexity and computational load increase

Engineering Contradiction:
Improvecompression performanceVSAvoidencoding complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the encoding process into distinct stages: evaluating different region partitioning options, selecting the optimal configuration, and executing the transform. This segmentation allows the system to manage complexity by breaking down the decision-making process into manageable steps, where each stage handles a specific aspect of the encoding task.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system evaluates multiple region partitioning configurations but does not necessarily implement all of them. Instead, it selects the most suitable configuration for each specific data block, performing partial actions (evaluating options) without committing to excessive complexity (implementing all configurations).

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If geometric partitioning with adaptive regions is implemented, then the transform efficiency is improved, but the processing time and computational resources increase

Engineering Contradiction:
Improvetransform efficiencyVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary evaluation of different region partitioning configurations before executing the actual transform. By pre-assessing the suitability of different partitioning schemes based on data characteristics, the system avoids unnecessary computational work and selects the most efficient configuration in advance, reducing overall processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The encoding system performs self-optimization by automatically selecting the appropriate number of regions based on the inherent characteristics of the input data. This self-service mechanism eliminates the need for external intervention or manual configuration, allowing the system to adaptively optimize its own performance without additional overhead.

Inventive Principle:
Principle #25Self-service

4Ease of manufacture

If a fixed transform block size is used, then the implementation is simple and hardware-friendly, but the adaptability to different data characteristics is limited

Engineering Contradiction:
Improveimplementation simplicityVSAvoidadaptability to data characteristics
Core Design Contradiction:
Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal encoding framework that can handle multiple data characteristics and block types within a single implementation. The adaptive region partitioning system is designed to work with various input configurations, making the encoder versatile enough to handle different data types and sizes without requiring separate specialized implementations for each case.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3918784B1Shape adaptive discrete cosine transform for geometric partitioning with an adaptive number of regions
Publication Date: 2026.05.20 DOLBY INTERNATIONAL AB
  • EP3918784B1 patent drawingFigure 1
  • EP3918784B1 patent drawingFigure 2
  • EP3918784B1 patent drawingFigure 3

AI summary

A decoder includes circuitry configured to receive a bitstream; determine a first region, a second region, and a third region of a current block and according to a geometric partitioning mode and decoding the current block. Related apparatus, systems, techniques and articles are also described. Decoder may determine, from a signal contained in the bitstream, a coding transformation type to decode the first region, the second region, and/or the third region, the coding transformation type characterizing at least an inverse block discrete cosine transformation and an inverse shape adaptive discrete cosine transformation, and the decoding of the current block may include using the determined transformation type for inverse transformation for each of the first region, the second region and/or the third region