Two-Level Quantization in Video Coding Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently managing quantization steps and rounding offsets, leading to suboptimal video quality and bitrate efficiency, particularly in handling varying luma sample ranges and coefficient positions within video blocks.
Innovation Solution
The implementation of a multi-level quantizer that selectively enables first and second level quantizations based on specific rules, such as quantization precision and coefficient counts, allowing for adaptive quantization steps and rounding offsets to be signaled in the bitstream, enabling finer control over quantization processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single quantization precision is used for all video blocks, then the device complexity is reduced, but the video quality and bitrate efficiency deteriorate due to inability to adapt to varying luma sample ranges and coefficient positions
Solution Approach 1:
The quantization process is divided into multiple levels with different quantization precisions. The patent applies a first quantization precision for certain coefficient positions and a second quantization precision for other positions, allowing adaptation to varying luma sample ranges without requiring a completely different quantization system for each case.
Solution Approach 2:
Different quantization precisions are applied to different coefficient positions within the video block. The patent specifies that a first quantization precision is used for coefficients at certain positions while a second quantization precision is used for coefficients at other positions, optimizing quality locally rather than uniformly across the entire block.
2Manufacturing precision
If adaptive quantization with multiple precisions is implemented, then video quality and bitrate efficiency improve, but the device complexity and processing overhead increase
Solution Approach 1:
The quantization precision is made dynamic by selectively applying different quantization precisions based on coefficient position and luma sample range characteristics. The patent enables the quantizer to adapt its precision level dynamically rather than using a fixed precision, improving quality where needed while maintaining efficiency elsewhere.
Solution Approach 2:
The patent changes the quantization parameter (precision level) based on the characteristics of the video block, specifically the luma sample range and coefficient position. By adjusting the quantization parameter according to local characteristics, the system achieves higher precision where required while avoiding unnecessary complexity in regions where it is not needed.
3Manufacturing precision
If quantization step is smaller than 2^1/6, then the manufacturing precision of quantization improves, but the bitrate increases due to more bits required to represent the finer quantization levels
Solution Approach 1:
High precision quantization (smaller quantization step) is applied only to specific coefficient positions where it provides the most benefit, rather than uniformly across all coefficients. This local application of high precision reduces the overall bitrate increase while maintaining quality where it matters most.
Solution Approach 2:
The patent applies refined quantization (smaller quantization step) partially rather than excessively. By selecting specific coefficient positions for refined quantization based on their importance and characteristics, the system achieves sufficient precision without applying excessive quantization refinement that would unnecessarily increase bitrate.
Data Source
AI summary
The application relates to refined quantization steps in video coding. A method for video processing includes determining, during a conversion between a current video block and a bitstream representation of a video including the current video block, two-level quantization parameters including a first level quantization parameter QP1 and a second level quantization parameter QP2 for the current video block. The method further includes performing, at least based on the two-level quantization parameters, a first level quantization and/or a second level quantization during the conversion. The two-level quantization parameters correspond to different quantization precisions, respectively.