Secondary Transform Zero-Out for Video Coefficient Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in representing less likely intra prediction directions and motion vector directions, leading to increased bit usage and reduced compression efficiency.
Innovation Solution
Implementing a secondary transform and zero-out method for transform coefficients, where coefficients outside a defined region are set to zero, and determining flags based on neighboring coefficients within a sub-region to optimize entropy coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional intra coding without prediction is used, then the coding process is simple, but the compression efficiency is poor and bitrate requirements are high
Solution Approach 1:
The patent applies preliminary action by performing intra prediction on the current block before transform coding. The predictor block is generated from reference samples surrounding the current block, and this prediction is performed in advance to reduce the residual energy that needs to be coded, thereby improving compression efficiency while maintaining a manageable coding process
Solution Approach 2:
The patent introduces an intermediary element - the predictor block - which is generated from reference samples and used to predict the current block. This intermediary predictor block serves as a mediator between the reference samples and the transform coding process, enabling better compression by reducing the residual signal energy
2Measurement precision
If more intra prediction directions are supported to represent less likely directions, then the coding accuracy improves, but the number of bits required increases
Solution Approach 1:
The patent applies partial action by selectively applying the secondary transform only to specific coefficient groups that are likely to contain non-zero coefficients, rather than transforming all coefficient groups. This is controlled by flags that indicate whether secondary transform should be applied to each coefficient group, thereby reducing bit usage while maintaining prediction accuracy for the most important directions
Solution Approach 2:
The patent applies local quality by allowing different coefficient groups to have different transform applications based on their local characteristics. Each coefficient group can be independently flagged to receive secondary transform or not, enabling the coding process to adapt to local patterns in the residual signal and reduce overall bit usage
3Productivity
If secondary transform is applied to all coefficient groups, then the compression efficiency improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies partial action by selectively applying secondary transform only to coefficient groups that are likely to benefit from it, as indicated by flags. This selective application reduces the total number of transform operations performed, thereby reducing processing time while maintaining compression efficiency for the most beneficial cases
Solution Approach 2:
The patent changes the parameter of transform application from a fixed all-or-nothing approach to a selective approach controlled by flags. These flags act as parameters that dynamically control whether secondary transform is applied to each coefficient group, allowing the system to balance compression efficiency and processing time based on the actual content characteristics
Data Source
AI summary
A non-transitory computer-readable storage medium stores instructions which, when executed by a processor, cause the processor to perform an encoding method. In the method, when a first region of a transform block (TB) includes a first sub-region and a second sub-region with transform coefficients calculated by a secondary transform, whether the secondary transform is performed on the first region is determined. The transform coefficients of the second sub-region are set to zero. When the secondary transform is determined to be performed on the first region that includes the second sub-region with the transform coefficients set to 0, transformation coefficients of a second region in the TB are determined as zero. The second region is outside the first region. Information of the TB is encoded into a video bitstream. The video bitstream is transmitted.