Video Frame Encoding Unit Division Method and Apparatus

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video frame coding methods result in relatively low coding speed due to inefficient division of coding units, particularly influenced by frame types and sizes, leading to error accumulation and reduced coding accuracy.

Innovation Solution

The method determines specific division depths for different frame types and coding unit sizes by establishing corresponding coding unit types, allowing for tailored division operations that avoid unnecessary depth divisions and error diffusion, thereby improving coding speed and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If all coding units are divided to the lowest level through layer-by-layer comparison, then coding accuracy is improved, but coding speed deteriorates

Engineering Contradiction:
Improvecoding accuracyVSAvoidcoding speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent applies different division depths to different coding units based on their specific characteristics (frame type, coding unit type, and size). Instead of uniformly dividing all coding units to the lowest level, the method selectively divides only those coding units that require higher accuracy, thereby improving coding speed while maintaining necessary coding accuracy.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the division depth parameter dynamically based on frame type (I-frame, P-frame, B-frame) and coding unit type. By adjusting this parameter according to different scenarios, the method avoids unnecessary deep division in frames or regions where it is not needed, thus improving coding speed without significantly compromising coding accuracy.

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If coding units with different sizes are uniformly divided, then processing simplicity is improved, but error accumulation increases

Engineering Contradiction:
Improveprocessing simplicityVSAvoiderror accumulation
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent determines different target division depths for coding units of different sizes based on their specific characteristics. Larger coding units may require deeper division to capture fine details, while smaller coding units can be divided to shallower levels. This localized approach prevents error accumulation by avoiding over-division of small units while ensuring adequate division of large units.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent applies partial division action by determining target division depths that are sufficient for each coding unit's needs rather than applying maximum division uniformly. This prevents excessive division of small coding units that would lead to error accumulation, while still applying adequate division where necessary.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3793196B1Video frame encoding unit division method and apparatus, and storage medium and electronic apparatus
Publication Date: 2024.01.10 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • EP3793196B1 patent drawingFigure 1~2
  • EP3793196B1 patent drawingFigure 3~4
  • EP3793196B1 patent drawingFigure 5~6

AI summary

Disclosed are a video frame encoding unit division method and apparatus, and a storage medium and an electronic apparatus. The method comprises: according to frame types and encoding unit types that have a correlation, determining a target encoding unit type corresponding to a target frame type of a target frame, wherein the target encoding unit type is used for indicating a division depth when the target frame is being divided; when a target encoding unit of the target encoding unit type is divided in the target frame, determining, according to encoding unit information of the target encoding unit, whether the target encoding unit satisfies a target condition, so as to obtain a target result; and executing a division operation corresponding to the target result on the target encoding unit. The present application solves the technical problem of a relatively low frame encoding speed in the related art.