Context Modeling for Multiple Transform Matrix Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards face challenges in efficiently managing bandwidth demand due to high data requirements for digital video, particularly with the increasing number of connected devices, and existing multiple transform selection (MTS) technologies are sub-optimal for color-rich sequences and lack separate control for intra block copy (IBC) mode.
Innovation Solution
The implementation of context modeling and multiple transform matrix selection methods that allow for selective application of MTS operations based on coding modes, block shapes, and color formats, enabling the use of predefined transforms for prediction errors during video processing, and signaling MTS information at various video data units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple transform selection (MTS) operation is applied to all video blocks, then video coding efficiency is improved, but device complexity increases due to need for separate control and signaling
Solution Approach 1:
The patent applies different MTS control strategies to different video block types. Specifically, MTS is enabled for inter-mode blocks but disabled for intra-mode blocks (including IBC mode). This local differentiation allows the system to leverage MTS benefits where most effective (inter-mode with motion compensation) while avoiding unnecessary complexity in other modes where traditional transforms suffice.
Solution Approach 2:
The patent segments the video processing workflow by separating MTS application based on coding mode. By dividing blocks into inter-mode and intra-mode categories, and further subdividing intra-mode into conventional intra and IBC modes, the system can selectively apply MTS only where beneficial, reducing overall system complexity while maintaining coding efficiency gains.
2Measurement precision
If MTS operation is applied to color-rich sequences, then coding precision is improved, but bandwidth demand increases
Solution Approach 1:
The patent changes the parameter of transform selection based on sequence characteristics. By detecting whether a sequence is color-rich and adjusting MTS application accordingly, the system can maintain high coding precision for color-rich content while reducing bandwidth consumption for sequences where MTS would provide minimal benefit but increase bitstream data requirements.
3Adaptability or versatility
If separate control for IBC mode is implemented, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent merges the control logic for IBC mode with the existing intra-mode control structure. By treating IBC mode similarly to conventional intra-mode in terms of MTS disablement, the system achieves separate control capability without requiring entirely new control pathways, thus improving adaptability while minimizing additional complexity.
Data Source
AI summary
Devices, systems and methods for digital video coding, which includes context modeling and multiple transform matrix selection, are described. An exemplary method for video processing includes determining, for a conversion between a current video block of a video and a bitstream representation of the video, that an indication of a transform type in a multiple transform selection (MTS) operation is excluded from the bitstream representation, and wherein the current video block is coded using an intra block copy mode, and performing the conversion, wherein the indication is in a sequence parameter set, a picture parameter set, a tile group header, a slice header, a picture header, a tile header, a coding tree unit row, or a video data unit, and wherein the MTS operation includes using a transform from a plurality of predefined transform candidates for transforming a prediction error of the current video block during the conversion.