Transform Skip Coefficient Coding With Dynamic Block Size Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face inefficiencies in handling intra block copy (IBC) and transform skip modes, particularly for screen content, due to issues like reference block overlap and transform inefficiencies, leading to increased bandwidth demand and coding complexity.
Innovation Solution
Implement techniques such as adaptive motion vector resolution (AMVR), quantized residual block differential pulse-code modulation (QR-BDPCM), and improved coefficient coding in transform skip mode, including determining maximum allowed dimensions and selective application of transform skip modes, to enhance coding efficiency and reduce bandwidth requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transform skip mode is applied with fixed maximal dimensions in current video coding standards, then coding simplicity is maintained, but coding efficiency is limited due to inflexibility in adapting to different prediction modes and block sizes
Solution Approach 1:
The patent makes the maximal dimension for transform skip mode dynamic by deriving it from the intra prediction mode of the current block. Different prediction modes (e.g., planar, angular modes 0-34) map to different maximal dimensions (e.g., 64, 32, 16, or 8), allowing the system to adapt the transform skip block size according to the prediction characteristics rather than using a fixed dimension for all cases.
Solution Approach 2:
The patent changes the parameter of maximal transform skip dimension based on the intra prediction mode parameter. By linking the dimension parameter to the prediction mode, the system optimizes coding efficiency for different types of content characteristics without requiring complex additional signaling, thus improving productivity while controlling complexity.
2Measurement precision
If coded block flag signaling is applied for each coefficient group in transform skip mode, then coding precision is improved, but overhead is increased
Solution Approach 1:
The patent applies coded block flag signaling partially - only for coefficient groups that exceed the derived maximal dimension for the current block's prediction mode. For coefficient groups within the maximal dimension, the signaling is omitted. This partial application maintains sufficient coding precision for significant residual regions while reducing overall overhead by avoiding redundant signaling for smaller blocks.
3Adaptability or versatility
If transform skip mode supports only luma component in current standards, then implementation complexity is reduced, but adaptability is limited for chroma transform skip
Solution Approach 1:
The patent extends transform skip mode to be universal across both luma and chroma components. The same intra prediction mode-based maximal dimension derivation and coded block flag signaling mechanisms are applied to chroma blocks, allowing chroma transform skip to function similarly to luma transform skip. This multi-functionality improves adaptability without significantly increasing implementation complexity since the underlying logic remains consistent.
Data Source
AI summary
Devices, systems and methods for coefficient coding in transform skip mode are described. An exemplary method for video processing includes determining, for encoding one or more video blocks in a video region of a visual media data into a bitstream representation of the visual media data, a maximum allowed dimension up to which a current video block of the one or more video blocks is allowed to be encoded using a transform skip mode such that a residual of a prediction error between the current video block and a reference video block is represented in the bitstream representation without applying a transform; and including a syntax element indicative of the maximum allowed dimension in the bitstream representation.