Video Encoding Hierarchical Coding Units

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional video codecs face inefficiencies in encoding and decoding high-resolution video content due to fixed macroblock sizes, leading to increased macroblocks per picture and decreased data compression efficiency.

Innovation Solution

The method involves adaptive determination of coding unit size and encoding mode based on image characteristics, using hierarchical encoding units of varying sizes to split images into maximum, deeper, and minimum coding units, and employing Rate-Distortion Optimization for optimal encoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If fixed macroblock sizes are used for encoding, then encoding simplicity is maintained, but data compression efficiency decreases for high-resolution video

Engineering Contradiction:
Improveencoding simplicityVSAvoiddata compression efficiency
Core Design Contradiction:
Ease of manufactureVSProductivity

Solution Approach 1:

The patent applies dynamic coding unit sizes that adapt to different regions of the video frame and different depth levels in the hierarchical structure. Instead of fixed macroblock sizes, the encoding system dynamically selects coding unit sizes (e.g., 64x64, 32x32, 16x16, 8x8 blocks) based on local image characteristics and depth, allowing larger units for simple regions and smaller units for complex regions, thereby improving compression efficiency while maintaining encoding feasibility

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements local quality adaptation by applying different coding unit sizes and encoding modes to different spatial regions and depth levels. Each coding unit can be independently configured with appropriate partitioning and prediction modes based on local image complexity, motion characteristics, and depth information, enabling optimized compression for each local region rather than uniform encoding across the entire frame

Inventive Principle:
Principle #3Local quality

2Productivity

If hierarchical encoding units of varying sizes are used, then data compression efficiency improves, but device complexity increases

Engineering Contradiction:
Improvedata compression efficiencyVSAvoidencoding apparatus complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the video encoding process into a hierarchical structure with multiple depth levels, where each level divides coding units into smaller partitions. This segmentation allows systematic exploration of different coding unit sizes (from large 64x64 blocks down to small 8x8 blocks) in an organized manner, managing the complexity through structured division rather than random trial of all possible configurations

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs dynamic selection of coding unit sizes and partitioning modes at each depth level based on local image characteristics. The encoding apparatus adaptively determines the optimal coding unit size for each region and depth, avoiding the need to evaluate all possible configurations exhaustively. This dynamic adaptation reduces computational complexity while maintaining high compression efficiency through intelligent local optimization

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If the number of macroblocks per picture increases, then image detail representation improves, but encoding and decoding time increases

Engineering Contradiction:
Improveimage detail representationVSAvoidencoding and decoding time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent segments the image into a hierarchical structure where detail representation is achieved through progressive refinement across multiple depth levels rather than using a large number of uniformly small macroblocks. This segmentation allows the system to use larger coding units for regions where high detail is not critical and smaller units only where necessary, reducing the total number of coding units compared to conventional fixed-size macroblock approaches while maintaining image detail quality

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality optimization by using smaller coding units and more sophisticated partitioning only in regions requiring high detail representation (such as edges, textures, or motion boundaries), while using larger coding units in smooth or homogeneous regions. This local adaptation reduces the overall number of coding units and processing operations compared to uniformly fine-grained macroblocks, thereby reducing encoding and decoding time while preserving necessary image detail

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3267683B1Method and apparatus for encoding video
Publication Date: 2018.12.05 SAMSUNG ELECTRONICS CO LTD
  • EP3267683B1 patent drawingFigure 1~3
  • EP3267683B1 patent drawingFigure 4
  • EP3267683B1 patent drawingFigure 5

AI summary

Disclosed is a method of encoding a video, the method including: splitting a current picture into at least one maximum coding unit; determining a coded depth to output a final encoding result according to at least one split region obtained by splitting a region of the maximum coding unit according to depths, by encoding the at least one split region, based on a depth that deepens in proportion to the number of times the region of the maximum coding unit is split; and outputting image data constituting the final encoding result according to the at least one split region, and encoding information about the coded depth and a prediction mode, according to the at least one maximum coding unit.