Chrominance Transform Selection for Video Encoding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current VVC standards limit encoding efficiency for chrominance signals by only applying the default transform mode regardless of signal format, particularly in formats like 4:4:4 and 4:2:2 where chrominance signals have a higher information amount and resolution similar to luminance signals, leading to suboptimal energy concentration and encoding performance.
Innovation Solution
An encoding and decoding device that selectively applies transform types such as DCT2, DCT8, and DST7 to chrominance signals based on the image signal format, allowing for improved energy concentration and encoding efficiency by determining the best transform combinations for each chrominance block in 4:4:4, 4:2:2, and 4:2:0 formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the default transform mode (DCT2) is applied to chrominance signals in all signal formats, then the encoding device maintains simplicity in transform selection, but the encoding efficiency and energy concentration for chrominance signals in high-resolution formats (4:4:4, 4:2:2) are suboptimal
Solution Approach 1:
The patent changes the transform type parameter dynamically based on the signal format. For 4:2:0 format, DCT2 is used; for 4:2:2 format, DST7 is used; for 4:4:4 format, DCT8 is used. This parameter adaptation resolves the contradiction by optimizing encoding efficiency for each signal format while maintaining a manageable transform selection framework.
Solution Approach 2:
The transform selection is made dynamic rather than static. The encoding device automatically selects different transform types based on the detected signal format, allowing the system to adapt to different chrominance signal characteristics without manual intervention, thereby improving encoding efficiency across various formats.
2Productivity
If multiple transform types (DCT8, DST7) are selectively applied to chrominance signals based on signal format, then the encoding efficiency and energy concentration are improved, but the device complexity increases
Solution Approach 1:
The patent manages transform selection complexity by organizing transforms into a structured parameter system where the signal format determines the transform type. This parameter-based approach simplifies the control logic despite having multiple transform types, as the selection rule is clearly defined based on signal format characteristics.
Solution Approach 2:
The patent segments the transform selection process into distinct categories based on signal formats (4:2:0, 4:2:2, 4:4:4). Each segment has a predetermined optimal transform type, which simplifies the overall selection process by breaking down the complex decision into manageable format-specific rules.
3Device complexity
If the transform type is fixed for chrominance signals regardless of signal format, then the device complexity is reduced, but the energy concentration and encoding performance for high-resolution chrominance signals are suboptimal
Solution Approach 1:
The patent adjusts the transform type parameter according to the signal format to achieve optimal energy concentration. For high-resolution chrominance signals in 4:2:2 and 4:4:4 formats, appropriate transforms (DST7, DCT8) are selected to concentrate energy effectively, thereby improving encoding precision without excessive complexity.
Data Source
AI summary
An encoding device performing an encoding process in units of blocks into which images constituted by luminance signals and chrominance signals are divided includes: a predictor configured to predict a target chrominance block which is an encoding target block of the chrominance signals and generate a prediction chrominance block; a residual generator configured to generate a chrominance prediction residual indicating a difference between the target chrominance block and the prediction chrominance block; and a transformer configured to apply a transform process to the chrominance prediction residual and generate transform coefficients, wherein the transformer is configured to control, according to a signal format of the images, a transform type to be applied to the transform process.