Matrix-Based Intra Prediction with Unified Offset Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing Matrix-based Intra Prediction (MIP) technology in video coding requires significant storage space and prolongs processing time due to varying parameters for luma blocks of different sizes, affecting coding and decoding efficiency.
Innovation Solution
Implementing a uniform offset parameter to unify the right shift parameters for different sizes and MIP mode numbers, reducing the need for table lookups and storage of multiple parameter sets, and optimizing the coding and decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If MIP mode is adopted for luma prediction with different block sizes, then prediction accuracy is improved, but storage space requirement increases and processing time is prolonged
Solution Approach 1:
The patent changes the parameter representation by introducing a unified offset parameter that replaces multiple size-specific right shift parameters. This parameter transformation allows the system to maintain different prediction behaviors for different block sizes while using a single standardized parameter set, thereby reducing storage requirements without sacrificing prediction accuracy.
Solution Approach 2:
The unified offset parameter serves multiple functions simultaneously: it works for all block sizes (4x4, 8x8, 16x16, etc.) and all MIP modes (0-34), replacing what would otherwise require separate parameter sets for each combination. This universal parameter enables the same parameter to be reused across different contexts, reducing overall storage space while maintaining the ability to handle diverse prediction scenarios.
2Measurement precision
If MIP mode is adopted for luma prediction with different block sizes, then prediction accuracy is improved, but processing time is prolonged
Solution Approach 1:
By transforming the parameter system to use a unified offset parameter with a fixed range (0-7), the patent enables more efficient parameter handling. The encoder can signal this compact parameter using fewer bits, and the decoder can quickly reconstruct the actual right shift value through a simple formula, reducing the time required for parameter transmission and processing compared to handling multiple separate parameter sets.
Solution Approach 2:
The patent performs preliminary definition of the unified offset parameter and its relationship to block size and MIP mode at the standard level. This pre-established framework allows encoders and decoders to quickly determine the appropriate right shift parameter without requiring complex lookups or calculations during actual prediction, thereby reducing processing time while maintaining prediction accuracy.
3Measurement precision
If separate right shift parameters are used for different block sizes and MIP modes, then prediction precision is maintained, but device complexity increases
Solution Approach 1:
The patent introduces a unified offset parameter that fundamentally changes how right shift values are represented. Instead of storing and managing separate parameters for each block size and MIP mode combination, the system uses a single offset parameter (0-7) that works universally. This parameter transformation simplifies the device architecture by eliminating the need for complex parameter management structures while preserving the precision needed for accurate prediction.
Data Source
AI summary
Picture coding and decoding methods, an encoder, a decoder and a storage medium are provided. The encoder, before performing coding processing according to a matrix-based intra prediction (MIP) mode, sets initial right shift parameters corresponding to different sizes and different MIP mode numbers as an uniform offset parameter, the offset parameter indicating a number of right shifting bits of a predicted value, and when performing coding processing according to the MIP mode, performs coding processing according to the offset parameter. The decoder, before performing decoding processing according to an MIP mode, sets initial right shift parameters corresponding to different sizes and different MIP mode numbers as an uniform offset parameter, the offset parameter indicating a number of right shifting bits of a predicted value, and when performing decoding processing according to the MIP mode, performs decoding processing according to the offset parameter.