Adaptive Quantization Parameter Determination for HDR Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding and decoding methods fail to accurately reflect image characteristics, leading to suboptimal compression efficiency and quality, particularly in high dynamic range (HDR) images, due to the use of a single quantization step for all cases without considering statistical information of the current block or previously reconstructed samples.
Innovation Solution
An image encoding and decoding method that determines a differential quantization parameter based on statistical information of the current block or previously reconstructed samples, allowing for adaptive adjustment of quantization to better reflect image characteristics and improve compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single quantization step is used for all cases, then device complexity is reduced, but image quality and compression efficiency deteriorate
Solution Approach 1:
The patent applies local quality by determining quantization parameters differently for different blocks based on their statistical characteristics. Specifically, it calculates a first quantization parameter for blocks where the standard deviation of gradient magnitude is below a threshold, and a second quantization parameter for blocks where the standard deviation is at or above the threshold. This allows each block to be quantized according to its local image characteristics, improving overall image quality while maintaining reasonable complexity.
Solution Approach 2:
The patent changes the quantization parameter based on statistical information of the current block. It computes the gradient magnitude of the current block, calculates the standard deviation of these gradients, and uses this statistical measure to select appropriate quantization parameters. This dynamic parameter adjustment improves compression efficiency and image quality by adapting to local variations in image content.
2Ease of operation
If a single quantization step is used for all cases, then ease of operation is improved, but compression efficiency deteriorates
Solution Approach 1:
The patent dynamically changes quantization parameters based on statistical analysis of each block's gradient magnitude. By computing the standard deviation of gradients and comparing it to a threshold, the system selects from different quantization parameter sets (firstQP or secondQP). This adaptive approach significantly improves compression efficiency by matching quantization strength to local image complexity while maintaining operational simplicity through automated threshold-based decision logic.
3Manufacturing precision
If statistical information analysis is performed for each block, then image quality improves, but device complexity increases
Solution Approach 1:
The patent performs statistical analysis locally for each block by calculating the gradient magnitude and its standard deviation specific to that block. This localized approach allows the system to capture fine-grained image characteristics without requiring global analysis, improving image quality while keeping computational complexity manageable through block-wise independent processing.
Solution Approach 2:
The patent applies partial action by performing statistical analysis only to the extent necessary - calculating gradient magnitude and standard deviation for each block, then comparing to a threshold. It does not perform exhaustive analysis but rather the minimum necessary statistical computations to make informed quantization decisions, balancing image quality improvement with acceptable processing complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is an image encoding or decoding method and device capable of increasing image compression efficiency by determining a differential quantization parameter of a current block based on statistical information of original samples of the current block or samples reconstructed previously to the current block. The image decoding method includes receiving a bitstream, obtaining a predicted quantization parameter value and a quantized transformation coefficient of a current block from the bitstream, obtaining a differential quantization parameter of the current block based on statistical information of samples reconstructed previously to the current block, obtaining a quantization parameter of the current block based on the differential quantization parameter of the current block and the predicted quantization parameter value, and inversely quantizing the quantized transformation coefficient of the current block based on the quantization parameter of the current block.