Palette Coding Transpose Flag Signaling for Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing palette coding methods are inefficient when the palette size is zero or all pixels in a block are escape pixels, leading to redundancy and duplication in signaling, similar to Pulse-Code Modulation (PCM) mode, which hampers coding efficiency.
Innovation Solution
Conditional signaling of the palette transpose flag based on the maximum index value or palette size, with specific syntax elements and flag inferences to optimize palette coding, such as omitting unnecessary signaling and updating processes when palette size is zero or less than a threshold, and invoking PCM mode when palette size equals zero.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If palette coding is used with palette size equal to zero or all pixels being escape pixels, then the coding method can handle edge cases, but redundancy and duplication in signaling occur leading to reduced coding efficiency
Solution Approach 1:
The patent changes the parameter of palette size to zero to invoke PCM mode, transforming the coding approach from palette-based to direct pixel value coding. This parameter change allows the system to adapt to edge cases where palette coding would be inefficient, resolving the contradiction between adaptability and coding efficiency by dynamically switching coding modes based on content characteristics
2Loss of information
If conditional signaling of palette transpose flag is implemented, then unnecessary signaling is reduced, but complexity in determining when to signal increases
Solution Approach 1:
The patent implements dynamic signaling where the palette transpose flag is conditionally signaled based on the relationship between current and previous palette sizes. When the palette size changes, the flag is signaled; otherwise, it is inferred. This dynamic approach reduces redundant signaling while maintaining simplicity by using a clear conditional rule based on palette size comparison
3Measurement precision
If palette size is explicitly signaled, then decoder can accurately determine palette dimensions, but additional bits are required in the bitstream
Solution Approach 1:
The patent signals palette size information partially - only when necessary for decoder reconstruction. By using conditional signaling based on palette size changes and invoking PCM mode when palette size is zero, the patent provides exactly the right amount of information needed without excessive signaling, optimizing the balance between measurement precision and data volume
Data Source
Figure 1A~1B
Figure 1C
Figure 2
AI summary
Methods for coding a block of video data using palette coding in a video coding are disclosed. According to one embodiment, the palette transpose flag is signaled conditionally depending on the maximum index value or the palette size. If the maximum index value or the palette size is equal to or greater than a threshold size, the palette transpose flag is signaled at an encoder side or parsed at a decoder side. Otherwise, the palette transpose flag is not signaled or parsed. In another embodiment, the last palette and the last palette size are updated by the current palette with the current palette size conditionally. For example, updating the last palette and the last palette size by the current palette with the current palette size is skipped if the current coding unit is coded using a pulse-coded modulation (PCM) mode or the current palette size is zero.