Variable Tree-Structure Transformation Units for Video Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video codecs struggle with efficiently encoding and decoding high-resolution video content due to their reliance on fixed-sized macroblocks for transformation and inverse-transformation, which does not effectively adapt to varying image characteristics.

Innovation Solution

A video encoding method that uses transformation units in a variable tree-structure, determined from a base transformation unit based on a maximum split level, allowing for transformation and inverse-transformation with units of varying sizes and shapes, to efficiently encode and decode video content considering image characteristics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed-sized macroblocks are used for transformation, then device complexity is reduced and ease of operation is improved, but encoding efficiency and adaptability to image characteristics deteriorate

Engineering Contradiction:
Improveadaptability to image characteristicsVSAvoidcomplexity of transformation unit structure
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The current block is divided into multiple transformation units of different sizes (e.g., 4x4, 8x8, 16x16, 32x32) based on image characteristics. This segmentation allows the encoding system to adapt to varying frequency components in different regions of the image, improving encoding efficiency while maintaining manageable complexity through hierarchical division.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The transformation unit size is made dynamic rather than fixed. The encoder selectively determines the appropriate transformation unit size for each current block based on image characteristics such as gradient magnitude and variance. This dynamic adaptation allows the system to optimize encoding for both smooth and high-frequency regions without requiring excessive complexity.

Inventive Principle:
Principle #15Dynamics

2Productivity

If variable-sized transformation units are used, then encoding efficiency is improved, but calculation cost and processing time increase

Engineering Contradiction:
Improveencoding efficiencyVSAvoidcalculation cost
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The invention changes the parameter of transformation unit size based on image characteristics. By calculating metrics such as gradient magnitude and variance for each current block, the system dynamically selects the appropriate transformation unit size (4x4, 8x8, 16x16, or 32x32), optimizing encoding efficiency while controlling calculation cost through targeted analysis rather than exhaustive processing.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

Different transformation unit sizes are applied to different regions of the image based on local characteristics. High-frequency regions with large gradients use smaller transformation units, while smooth regions use larger units. This local adaptation improves overall encoding efficiency without requiring maximum computational resources for every block.

Inventive Principle:
Principle #3Local quality

3Adaptability or versatility

If multiple transformation unit sizes are supported, then adaptability to different image regions is improved, but information transmission overhead increases

Engineering Contradiction:
Improveadaptability to different image regionsVSAvoidtransmission overhead
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The transformation unit selection is segmented into discrete size categories (4x4, 8x8, 16x16, 32x32), allowing the system to convey size information efficiently through compact indexing rather than transmitting full dimensional data. This segmentation reduces transmission overhead while maintaining adaptability to different image regions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The encoder performs preliminary analysis of image characteristics (gradient magnitude, variance) before selecting transformation unit size. This preliminary action allows the system to make informed decisions about transformation unit selection, reducing the need for extensive post-processing information transmission and minimizing overhead.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3448027B1Video encoding method and apparatus using transformation unit of variable tree structure
Publication Date: 2020.11.04 SAMSUNG ELECTRONICS CO LTD
  • EP3448027B1 patent drawingFigure 1~2
  • EP3448027B1 patent drawingFigure 3
  • EP3448027B1 patent drawingFigure 4

AI summary

Provided is a video encoding method, in order to encode a current region of a video, performing transformation on the current region by using transformation units in a variable tree-structure which are determined from among transformation units that are hierarchically split from a base transformation unit with respect to the current region and which are generated based on a maximum split level of a transformation unit; and outputting encoded data of the current region, information about an encoding mode, and transformation-unit hierarchical-structure information comprising maximum size information and minimum size information of the transformation unit with respect to the video.