Image Block Quantization Parameter Selection for Video Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression technologies face challenges in achieving effective coding and decoding due to the loss of details in image blocks with different features during downsampling, leading to poor reconstructed image quality.

Innovation Solution

The method involves parsing a code stream to determine appropriate quantization parameters for image blocks based on their coding schemes, allowing for differential coding and reconstruction that accounts for the unique features of each block, thereby improving the coding and decoding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If downsampling coding scheme is used to compress image data, then compression efficiency is improved, but image quality deteriorates due to loss of details

Engineering Contradiction:
Improvecompression efficiencyVSAvoidimage quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The image is divided into multiple image blocks, and each block is independently classified into different categories (first category for smooth regions, second category for detailed regions). This segmentation allows different coding schemes to be applied to different regions, achieving both compression efficiency and image quality preservation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different quantization parameters are applied to different image blocks based on their category. Smooth regions use one quantization parameter while detailed regions use another, ensuring that each region is processed with the appropriate quality level. This local differentiation resolves the contradiction by maintaining high quality where needed while achieving compression in suitable areas.

Inventive Principle:
Principle #3Local quality

2Device complexity

If uniform quantization parameter is applied to all image blocks, then device complexity is reduced, but coding effectiveness deteriorates

Engineering Contradiction:
Improvecoding complexityVSAvoidcoding effectiveness
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The quantization parameter is made dynamic rather than static. The system dynamically selects different quantization parameters based on the classification results of image blocks. This dynamic adaptation improves coding effectiveness without significantly increasing device complexity, as the selection process follows clear classification rules.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Image blocks are classified into different categories before the actual coding process. This preliminary classification allows the system to prepare appropriate quantization parameters in advance, making the subsequent coding process more effective while keeping the overall complexity manageable through systematic organization.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3648462B1Image processing method and device
Publication Date: 2023.10.25 HUAWEI TECH CO LTD
  • EP3648462B1 patent drawingFigure 1~2
  • EP3648462B1 patent drawingFigure 3
  • EP3648462B1 patent drawingFigure 4

AI summary

This application provides an image processing method, device, and system. The method includes: parsing a code stream to obtain a first quantization parameter of a current image area in an original resolution coding scheme, a second quantization parameter of the current image area in a downsampling coding scheme, a coding scheme of a current to-be-reconstructed image block in the current image area, a quantization coefficient of the current to-be-reconstructed image block, and a prediction signal of the current to-be-reconstructed image block, where the coding scheme of the current to-be-reconstructed image block is the original resolution coding scheme or the downsampling coding scheme, and the current to-be-reconstructed image block is a part of the current image area; determining a quantization parameter of the current to-be-reconstructed image block according to the coding scheme of the current to-be-reconstructed image block; and reconstructing the current to-be-reconstructed image block based on the quantization parameter, the quantization coefficient, and the prediction signal of the current to-be-reconstructed image block, to obtain a current reconstructed image block. Therefore, a reconstructed image block obtained by a decoder side has a better effect.