Video Coding Unit Splitting Modes for Compression Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video compression technologies, such as HEVC and QTBT, face limitations in achieving optimal compression efficiency due to restricted coding unit splitting modes and redundancy in spatial topology, which hinder the representation of complex video data structures effectively.

Innovation Solution

The proposed method introduces a rich set of coding unit splitting modes, including binary and triple splits, with prioritization of triple splits over binary splits and asymmetric binary splits over symmetric splits, to create a flexible rectangular block-based representation in the compressed domain, ensuring no redundancy and maximizing reachable topologies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional HEVC quad-tree division is used, then the coding structure is simple, but the compression efficiency is limited due to restricted splitting modes

Engineering Contradiction:
Improvecompression efficiencyVSAvoidcoding structure complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides a coding block into multiple sub-blocks using binary splitting and triple splitting modes, creating a more granular segmentation structure. This allows the encoder to represent complex video content with greater precision, directly improving compression efficiency by better matching the block structure to the actual video data patterns.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a nested tree structure where binary trees and triple trees are combined hierarchically. The binary tree division can be further subdivided by triple tree division, creating a nested representation that allows multiple levels of splitting. This nested structure enables reaching more topologies within limited depth while maintaining an organized coding framework.

Inventive Principle:
Principle #7Nested doll (Nesting)

2Productivity

If more splitting modes are introduced to improve compression efficiency, then redundant spatial topologies increase

Engineering Contradiction:
Improvecompression efficiencyVSAvoidspatial topology redundancy
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent introduces asymmetric binary splitting modes where the split ratio is not necessarily equal (e.g., 1:3, 2:2, 3:1 divisions). This asymmetry allows the coding structure to better adapt to non-uniform content characteristics in video blocks, improving compression efficiency by creating more varied and appropriate block partitions without creating redundant symmetric topologies.

Inventive Principle:
Principle #4Asymmetry

Solution Approach 2:

The patent implements dynamic splitting mode selection where the encoder can choose between binary splitting, triple splitting, and their combinations based on the specific content characteristics. This dynamic approach allows the system to adapt to different video regions, applying the most efficient splitting mode locally while avoiding redundant representations across different regions.

Inventive Principle:
Principle #15Dynamics

3Device complexity

If the maximum tree depth is limited to reduce complexity, then the ability to represent complex video structures is reduced

Engineering Contradiction:
Improvetree depth complexityVSAvoidvideo structure representation capability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent adds a new dimension to the splitting structure by introducing triple tree division as an additional splitting operation beyond the traditional binary tree. This creates a multi-dimensional splitting capability where blocks can be divided through sequences of binary and triple splits, exponentially increasing the number of reachable topologies within the same depth limit and improving adaptability to complex video structures.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent creates a universal splitting framework that can handle various video content types using a combination of binary and triple splitting modes. This multi-functional approach allows the same coding structure to efficiently represent diverse video patterns (smooth regions, edges, textures, etc.) without requiring different dedicated structures, thereby improving versatility while maintaining manageable complexity.

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

Data Source

PatentEP3603066B1Multiple splits prioritizing for fast encoding
Publication Date: 2023.10.18 INTERDIGITAL VC HOLDINGS INC
  • EP3603066B1 patent drawingFigure 1
  • EP3603066B1 patent drawingFigure 2
  • EP3603066B1 patent drawingFigure 3

AI summary

A block of video data is split using at least one splitting rule and coded using existing transform sizes through one of several embodiments. In one embodiment, the block is split using a triple tree splitting operation if said at least two rectangular sub-blocks can be achieved with two successions of split operations with at least one triple tree split. In another embodiment, the video block is split using an asymmetric binary tree splitting if said at least two rectangular sub-blocks can be achieved with two successions of split operations with at least one asymmetric binary split. In another embodiment, a video block is split using successive splits, using both of the rules of the other embodiments.