Video Coding Quantization and De-quantization for Block-Adaptive Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques face challenges in efficiently compressing video data while maintaining video quality, particularly in handling various block sizes and transform types, which affects coding efficiency and visual quality.
Innovation Solution
The implementation of a quadtree with nested multi-type tree coding structure, combined with flexible luma and chroma block tree structures, multiple transform selection, and enhanced quantization and de-quantization processes, including scale and shift operations, to optimize video coding for various block sizes and transform types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional video coding techniques are used, then video data can be compressed, but coding efficiency and visual quality deteriorate when handling various block sizes and transform types
Solution Approach 1:
The patent implements dynamic quantization parameter adjustment based on block size and transform type. The system selects different quantization parameters (QP) for different block sizes (e.g., 4x4, 8x8, 16x16, 32x32) and transform types (DST, DCT), allowing the quantization process to adapt to local characteristics rather than using a fixed QP value. This dynamic adaptation improves both coding efficiency and visual quality by optimizing the trade-off between compression and quality for each specific case.
Solution Approach 2:
The patent changes multiple parameters simultaneously including quantization parameter (QP), transform type selection, and block size configuration. The system modifies these parameters based on content characteristics, enabling flexible optimization of compression performance. For example, different QP values are applied for different transform types (DST-VII, DCT-II) and block sizes, and the transform type itself is selected based on the block characteristics, achieving superior coding efficiency and visual quality.
2Loss of substance
If quantization parameter is increased to improve compression, then bit rate decreases, but video quality deteriorates
Solution Approach 1:
The patent applies different quantization parameters to different blocks based on their local characteristics such as block size, transform type, and content complexity. Instead of using a uniform QP across the entire video frame, the system selectively applies appropriate QP values to specific blocks. For example, larger QP values may be applied to blocks with less visual importance or simpler content, while smaller QP values are applied to blocks requiring higher quality, thereby achieving better overall compression efficiency without significant quality loss.
3Productivity
If multiple transform types are used to improve coding efficiency, then compression performance improves, but system complexity increases
Solution Approach 1:
The patent implements transform type selection based on block characteristics, dynamically changing the transform parameter (DST-VII, DCT-II, or other transforms) according to block size and content type. This allows the system to use multiple transform types to improve compression performance while managing complexity through systematic selection rules rather than exhaustive processing of all possible transforms for every block.
Data Source
AI summary
A method for video encoding is provided. The method may be applied in an encoder and includes: determining, for a first coding block, a quantization parameter and a quantized level; deriving a scale value by scaling the quantization parameter by a scale factor, determining a plurality of parameters associated with a coding unit (CU); obtaining a plurality of bit-shifts by bit-shifting the plurality of parameters; and obtaining a plurality of coefficients associated with the CU based on the quantized level, the scale value, and the plurality of bit-shifts.