Screen Content Coding with Adaptive Palette and Index Map
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current screen content coding techniques, such as intra block copy and color palette coding, face limitations in efficiency and complexity due to fixed block decomposition and difficulty in handling continuous color tones, especially in lossy coding modes.
Innovation Solution
The proposed solution involves an advanced screen content coding process using a palette table and index map coding, which includes pixel representation, palette table compression, color index map compression, string search, and residual compression, integrated with HEVC Range Extension to enhance coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If intra block copy with fixed block decomposition is used, then coding simplicity is maintained, but coding performance is bounded and cannot adapt to varying content characteristics
Solution Approach 1:
The patent implements dynamic block decomposition where the coding unit is divided into multiple prediction units with different sizes and types (inter prediction and intra prediction). The partitioning is adaptive based on content characteristics, allowing the encoder to select optimal block sizes and prediction modes for different regions, thus improving coding performance adaptability while managing complexity through structured decomposition
Solution Approach 2:
The coding unit is segmented into multiple prediction units (PUs) of varying sizes (e.g., 2Nx2N, 2NxN, Nx2N, NxN partitions). This segmentation allows different regions to be coded with appropriate prediction modes and block sizes, enabling the system to adapt to local content characteristics while maintaining overall coding efficiency
2Measurement precision
If block matching with motion estimation is performed, then coding accuracy is improved, but encoder complexity increases significantly in both computing and memory access
Solution Approach 1:
The patent applies different prediction qualities and methods to different regions. Inter prediction with motion compensation is used for regions with temporal redundancy, while intra prediction is used for regions with spatial redundancy. This local differentiation optimizes coding accuracy for each region type while reducing overall computational complexity by avoiding unnecessary complex operations in regions where simpler methods suffice
Solution Approach 2:
The patent performs block matching and motion estimation only for inter prediction modes when beneficial, rather than for all prediction units. The encoder selectively applies motion compensation based on rate-distortion optimization, performing partial action only where it provides sufficient gain to justify the computational cost, thus balancing accuracy improvement with complexity control
3Productivity
If color palette coding is used, then coding efficiency is improved for limited colors, but performance degrades for continuous or near-continuous color tones
Solution Approach 1:
The patent implements a universal prediction framework that can handle both discrete palette colors and continuous color tones through multiple prediction modes (inter prediction, intra prediction, and palette prediction). The encoder selects the appropriate mode based on content characteristics, making the system versatile for both screen content with limited colors and video content with continuous tones, thus achieving both coding efficiency and adaptability
4Measurement precision
If pseudo string match is applied, then lossless coding performance is maximized, but complexity overhead increases and lossy coding becomes difficult
Solution Approach 1:
The patent implements dynamic mode selection where the encoder can switch between lossless pseudo-string match coding and lossy prediction coding based on content characteristics and desired quality. For regions requiring lossless compression, pseudo-string match is applied; for other regions, more efficient lossy prediction modes are used, thus managing complexity overhead while maintaining high lossless performance where needed
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An apparatus is configured to perform a method for screen content encoding. The method includes deriving a palette table and a color index map based on a coding unit (CU). The method also includes encoding the palette table and encoding the color index map. The method further includes combining the encoded palette table and the encoded color index map for transmission to a receiver, where the palette table and index map are decoded to reconstruct a pixel block.