Point-Level Quantization Parameter Control for Video Coding Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution and high-frame-rate video poses challenges to display interfaces and links due to bandwidth limitations, necessitating improved encoding and decoding efficiency, particularly through better adaptation of quantization parameters (QP) to picture content.
Innovation Solution
A method and apparatus for encoding and decoding a region enhancement layer that adjusts quantization parameters (QP) at a point level, utilizing bit depth, picture complexity levels, and prediction residual values to enhance coding efficiency and precision, allowing for quantization at a low granularity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If quantization is performed at block level (CU level), then encoding complexity is reduced, but quantization precision and subjective quality are insufficient
Solution Approach 1:
The current invention segments the quantization process from block level to point level within a coding unit. Instead of applying a single QP to the entire CU, the patent divides the CU into multiple points (samples) and calculates individual QP values for each point based on local prediction residual characteristics. This segmentation enables finer-grained quantization control, improving quantization precision while maintaining manageable encoding complexity through efficient point-level operations.
Solution Approach 2:
The patent implements local quality by assigning different QP values to different points within the same CU based on their local characteristics. The QP calculation considers point-specific factors such as prediction residual magnitude and position, allowing important regions (e.g., flat areas with small residuals) to receive higher precision (lower QP) while less critical regions use coarser quantization (higher QP). This local adaptation improves overall subjective quality without uniformly increasing complexity across the entire block.
2Productivity
If QP is increased to reduce bitstream length, then compression efficiency improves, but picture quality deteriorates
Solution Approach 1:
The patent dynamically changes the QP parameter at point level rather than using a fixed or block-level QP. The QP value for each point is calculated based on local prediction residual characteristics, allowing the system to adaptively adjust quantization strength according to content importance. This parameter adaptation enables better quality-compression tradeoff by applying stronger compression only where appropriate while preserving quality in critical regions.
Solution Approach 2:
The invention introduces dynamics into the QP selection process by making QP values adaptive rather than static. Instead of using a uniform QP for the entire CU, the system calculates different QP values for different points based on their local characteristics and prediction residuals. This dynamic adaptation allows the encoding process to respond to local content requirements, improving both compression efficiency and picture quality simultaneously.
3Manufacturing precision
If point-level QP calculation is implemented, then subjective quality of flat content is improved, but encoding complexity increases
Solution Approach 1:
The patent segments the CU into individual points for separate QP calculation, enabling point-level quality control. This segmentation is implemented efficiently by processing points in a systematic order and reusing previously calculated values where possible, reducing the computational overhead compared to a naive implementation.
Solution Approach 2:
The QP calculation for each point utilizes locally available information (prediction residuals and neighboring point values) that is already present in the decoding process. The method serves itself by using the prediction error data that must be computed anyway for reconstruction purposes, avoiding the need for separate complex analysis passes and reducing overall encoding complexity.
Data Source
AI summary
A method comprises obtaining a bit depth of a current picture; obtaining a quantization parameter (QP) of a current coding unit (CU), where the current CU belongs to the current picture; obtaining a picture complexity level of the current CU; obtaining a prediction residual value of a current point, where the current point is located in a first row of the current CU, a prediction value of the current point is a reconstruction value of a reference point, and the reference point is located on an upper side of the current point; and obtaining a QP of the current point based on the bit depth of the current picture, the QP of the current CU, the picture complexity level of the current CU, and the prediction residual value of the current point.