Video Encoding Scaling List Selection for Adaptive Color Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The application of adaptive color transform (ACT) in video encoding can lead to visual deterioration in image quality due to the misapplication of scaling lists designed for luminance and chrominance signals, as luminance components are concentrated in the G component in the RGB space, causing inappropriate scaling lists to be applied to other components.
Innovation Solution
The encoding and decoding devices employ a configuration that determines a uniform scaling list for use when ACT is applied, ensuring consistent scaling across all components, thereby preventing visual deterioration when encoding and decoding videos with ACT.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If adaptive color transform (ACT) is applied to encoding-target blocks, then encoding efficiency is improved through color space transformation, but image quality deteriorates due to misapplication of scaling lists designed for luminance and chrominance signals
Solution Approach 1:
The patent applies different scaling list selection strategies to different components based on the ACT application status. When ACT is applied, a uniform scaling list is used for all components (Y, Cg, Co). When ACT is not applied, component-specific scaling lists are used. This local differentiation resolves the contradiction by adapting the scaling list application to the specific processing mode of each block.
Solution Approach 2:
The patent changes the scaling list parameter selection based on the ACT application flag. The quantization controller determines whether to apply a uniform scaling list or component-specific scaling lists by checking the ACT application status. This parameter change mechanism allows the system to switch between different scaling strategies to maintain image quality while preserving encoding efficiency benefits.
2Measurement precision
If component-specific scaling lists are used for luminance and chrominance signals, then quantization precision is improved, but visual deterioration occurs when ACT is applied because luminance components are concentrated in the G component causing inappropriate scaling list application
Solution Approach 1:
The patent dynamically changes the scaling list parameter based on the ACT application status. When ACT is applied, the system switches to using a uniform scaling list for all components, preventing the visual deterioration caused by inappropriate component-specific scaling. When ACT is not applied, component-specific scaling lists are used to maintain quantization precision. This parameter switching mechanism resolves the contradiction between precision and visual quality.
Solution Approach 2:
The patent prevents visual deterioration by proactively selecting appropriate scaling lists before quantization occurs. The quantization controller determines the scaling list selection based on the ACT application flag, ensuring that uniform scaling is applied to all components when ACT is active, thereby preventing the harmful effect of mismatched scaling lists before they can cause visual deterioration.
3Manufacturing precision
If different scaling lists are applied to different components in RGB space, then encoding precision is improved, but reliability decreases when ACT transforms RGB to YCgCo space because scaling lists designed for RGB components are inappropriate for YCgCo components
Solution Approach 1:
The patent changes the scaling list selection parameter based on the color space transformation status. When ACT transforms RGB to YCgCo space, the system switches to using a uniform scaling list that is appropriate for the transformed components. This parameter change ensures encoding reliability by preventing the application of RGB-designed scaling lists to YCgCo components, while still maintaining encoding precision through appropriate scaling when ACT is not applied.
Data Source
AI summary
An encoding device encodes each encoding-target block. The encoding device includes: a predictor configured to generate, for each of the components, a prediction block corresponding to the encoding-target block; a residual generator configured to generate, for each of the components, a prediction residual that represents a difference between the encoding-target block and the prediction block; a color space transformer configured to perform a color space transform process on the prediction residual of each of the components; a transformer configured to generate transform coefficients by performing a transform process on the prediction residual; a quantization controller configured to determine a scaling list to be used in a quantization process on the transform coefficients; and a quantizer configured to perform the quantization process on the transform coefficients by using the determined scaling list, wherein the quantization controller is configured to determine the scaling list, based on the color space transform process.