Palette Coding Transpose and Scanning Orders
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing palette coding techniques in HEVC struggle to further improve coding efficiency and reduce complexity, particularly in handling palette index prediction and scanning orders for screen content coding.
Innovation Solution
The introduction of multiple color index scanning orders, including non-horizontal scans like vertical, zig-zag, Hilbert, and diagonal scans, along with the option to transpose the color index map, allows for adaptive encoding and decoding of palette indices using context adaptive coding and additional prediction modes such as 'copy top-left', 'copy temporal', and 'column filling' modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple color index scanning orders and transpose operations are introduced, then coding efficiency is improved and bitrate is reduced, but device complexity increases
Solution Approach 1:
The patent introduces multiple scanning orders (horizontal, vertical, diagonal, zig-zag, Hilbert) and transpose operations that can be dynamically selected based on the characteristics of the current block. This dynamic adaptation allows the encoder to choose the most efficient scanning pattern for each block, improving coding efficiency while managing complexity through selective application
Solution Approach 2:
The patent changes the scanning parameters by introducing different scanning patterns and transpose operations. By varying the scanning order parameter according to block characteristics, the system achieves better compression ratios without permanently increasing complexity, as only the necessary transformations are applied
2Manufacturing precision
If multiple prediction modes are added, then coding precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the prediction process into multiple modes (copy left, copy top, copy top-left, column filling) that can be independently selected for different blocks. This segmentation allows the system to apply higher precision prediction only where needed, rather than uniformly across all blocks, thus improving overall coding precision while managing complexity
Solution Approach 2:
The patent implements partial application of complex prediction modes by allowing selective use of advanced prediction techniques only for blocks that benefit from them. Not all blocks require all prediction modes, so the system applies prediction complexity selectively, achieving high precision where necessary while maintaining lower complexity elsewhere
Data Source
AI summary
A method and apparatus for palette coding of a block of video data using multiple color index scanning orders involve: receiving input data associated with a current block; determining whether transposing a color index map for palette coding is selected; responsive to selecting transposing the color index map, encoding or decoding color indices of the current block according to a transposed palette or a transposed triplet palette; and responsive to not selecting transposing the color index map, encoding or decoding the color indices of the current block according to an original palette or an original triplet palette.


