Adaptive PQ Quantization for HDR Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing HDR video encoding methods based on Perceptual Quantizer (PQ) do not fully utilize quantization values and result in quantization loss due to fixed and large luminance ranges, failing to adapt to the actual luminance range of HDR videos.
Innovation Solution
An adaptive Perceptual Quantizer (PQ) method is introduced, where a quantization adjustment factor is determined based on the maximum and minimum luminance values of the video picture, allowing for adaptive encoding and decoding to optimize bit allocation and reduce quantization loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If uniform encoding is performed in a fixed and very huge luminance range using PQ, then the HDR video can be encoded to adapt to H.265/HEVC Main 10 Profile encoder, but the quantization value cannot be fully utilized and quantization loss occurs
Solution Approach 1:
The patent applies dynamics by transforming the fixed quantization parameter into an adaptive one. The quantization parameter is dynamically adjusted based on the actual luminance range of the HDR video content, allowing the encoding system to adapt its precision to the specific characteristics of each video segment, thereby fully utilizing available quantization values and reducing quantization loss
Solution Approach 2:
The patent implements parameter changes by modifying the quantization parameter (QP) based on the luminance range characteristics of the video content. Instead of using a fixed QP value, the system calculates and applies different QP values according to the actual luminance distribution, which optimizes the utilization of quantization values and improves encoding accuracy
2Device complexity
If a fixed quantization parameter is used for HDR video encoding, then the encoding process is simple, but the quantization loss increases due to inability to adapt to actual luminance range
Solution Approach 1:
The patent applies feedback by analyzing the actual luminance range of the HDR video content and using this information to adjust the quantization parameter. The system measures the luminance characteristics, feeds this information back to the quantization process, and accordingly optimizes the QP value to minimize quantization loss while maintaining encoding efficiency
Solution Approach 2:
The patent implements preliminary action by performing luminance range analysis before the actual encoding process. This preliminary analysis allows the system to determine the appropriate quantization parameter in advance, ensuring that the encoding process uses optimal quantization values from the start, thereby reducing quantization loss without significantly increasing overall complexity
Data Source
AI summary
Provided are picture encoding and decoding methods and apparatuses, and a picture encoding and decoding system. The picture encoding method includes: determining an adjustment factor according to pixel sampling values of a video picture; converting the video picture according to the adjustment factor, and encoding the converted video picture; and writing the adjustment factor into a bitstream obtained by encoding the converted video picture.