Palette Mode Coding for Chroma Subsampling Formats
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional video coding techniques are inefficient for compressing screen-generated content, such as remote desktops and wireless displays, which have discrete-tone characteristics and sharp lines, as they are optimized for continuous-tone natural images.
Innovation Solution
The implementation of palette-based video coding techniques, where a palette table with three color values is derived for each block, allowing for efficient encoding and decoding of pixels with either all three color values or a single luma value, depending on the presence of chroma components, to accommodate non-4:4:4 sampling formats like 4:2:2 and 4:2:0.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional video coding techniques are used, then continuous-tone natural images are compressed efficiently, but screen-generated content with discrete-tone characteristics and sharp lines is compressed inefficiently
Solution Approach 1:
The patent changes the fundamental parameters of video coding by introducing palette-based coding with discrete color values instead of continuous-tone representation. This allows the system to adapt to screen-generated content characteristics by using a limited palette of color values, thereby improving compression efficiency for discrete-tone content while maintaining compatibility with traditional coding approaches for natural images
Solution Approach 2:
The patent applies different coding strategies to different regions or types of content within the video stream. By identifying screen-generated content regions and applying palette-based coding specifically to those regions, the system achieves local optimization for discrete-tone content while preserving efficient coding for other content types, thus resolving the contradiction between compression efficiency and adaptability
2Productivity
If a single palette table with three color values is derived for each block, then coding efficiency is enhanced by adaptively using color values based on chroma component presence, but device complexity increases due to palette table derivation and conditional processing
Solution Approach 1:
The patent performs preliminary derivation of the palette table during the encoding phase, so that the decoder can simply look up color values from the pre-computed palette table. This shifts the computational complexity to the encoder side, allowing the decoder to maintain lower complexity while still achieving improved coding efficiency through adaptive use of color values based on chroma component presence
Solution Approach 2:
The patent introduces dynamic adaptation in the palette table usage by conditionally selecting the number of color values to retrieve based on the presence of chroma components. This dynamic behavior allows the system to optimize coding efficiency for different pixel types (with or without chroma) while managing complexity through systematic conditional logic rather than overly complex processing
3Quantity of substance
If chroma components are subsampled to reduce resolution, then data transmission efficiency is improved, but color accuracy and image quality deteriorate
Solution Approach 1:
The patent changes the approach to chroma subsampling by using palette-based coding where chroma information is represented through discrete color values in the palette table rather than continuous chroma components. This allows for more efficient subsampling while maintaining perceived color accuracy, as the palette entries are specifically chosen to represent the discrete-tone characteristics of screen-generated content
Solution Approach 2:
The patent uses a single palette table to serve multiple purposes - it stores color information for both luma-only pixels and pixels with chroma components. By copying the same palette table structure for different pixel types and conditionally retrieving appropriate color values, the system achieves efficient data representation without sacrificing color accuracy for subsampled chroma
Data Source
AI summary
Techniques are described to extend palette-mode coding techniques to cases where chroma components are at a different resolution than luma components. The entries of the palette table includes three color values and the three color values or a single one of the three color values are selected based on whether a pixel includes both a luma component and chroma components or only a luma component.


