Neighbor-Based Quantization Parameter Prediction for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding methods struggle to maintain transparent visual quality in high bit rate applications, particularly in smooth regions where loss of details is more noticeable due to the texture masking property of the human visual system, and they incur high overhead costs in signaling quantization parameters.
Innovation Solution
Determine quantization parameter predictors using multiple neighboring quantization parameters, reducing overhead by encoding the difference between the current quantization parameter and the predictor, and applying this method at various image partitions such as coding units, prediction units, or transform units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If quantization parameters are signaled at the slice or macroblock level, then quality control is achieved, but signaling overhead increases
Solution Approach 1:
The patent segments the picture into multiple coding units (CU, PU, or TU) and applies different quantization parameters to each segment. This allows fine-grained quality control while reducing overhead by predicting QP values for most segments based on neighboring segments, signaling only the differences rather than absolute values for each segment.
Solution Approach 2:
The patent enables local quality adjustment by allowing different quantization parameters for different coding units within the same picture. The QP can be adjusted based on the importance and visual characteristics of specific regions, enabling transparent quality in high bit rate applications while maintaining efficiency through prediction for less important regions.
2Reliability
If higher bit rate is allocated to smooth regions, then visual quality improves, but bit allocation complexity increases
Solution Approach 1:
The patent changes the quantization parameter (QP) based on the visual characteristics of different regions. Smooth regions receive higher QP values (lower quality) while textured regions receive lower QP values (higher quality), utilizing the texture masking property of the human visual system. This parameter adjustment is implemented efficiently through prediction mechanisms that reduce the complexity of bit allocation decisions.
Data Source
AI summary
Methods and apparatus are provided for determining quantization parameter predictors from a plurality of neighboring quantization parameters. An apparatus includes an encoder for encoding image data for at least a portion of a picture using a quantization parameter predictor for a current quantization parameter to be applied to the image data. The quantization parameter predictor is determined using multiple quantization parameters from previously coded neighboring portions. A difference between the current quantization parameter and the quantization parameter predictor is encoded for signaling to a corresponding decoder.