Video Coding Intra Prediction Matrix Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently managing bandwidth demand for digital video, particularly as the number of connected devices increases, leading to growing bandwidth requirements.
Innovation Solution
The proposed solution involves using a secondary transform during the decoding or encoding of video or images, specifically by determining the appropriate transform for a block of video data based on its representation in a bitstream, and applying techniques such as Matrix-based Intra Prediction (MIP), Intra Subblock Partitioning (ISP), and Affine Linear Weighted Intra Prediction (ALWIP) to enhance coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple intra coding methods (MIP, ISP, ALWIP) are integrated into the video coding system, then video coding efficiency and prediction accuracy are improved, but device complexity and processing overhead increase
Solution Approach 1:
The video block is divided into multiple sub-partitions using Intra Subblock Partitioning (ISP), where each sub-partition can be independently coded with different prediction methods. This segmentation allows the system to apply complex coding techniques selectively to specific regions, improving overall coding efficiency while managing complexity through localized processing.
Solution Approach 2:
The system dynamically selects and switches between multiple intra coding methods (MIP, ISP, ALWIP) based on the characteristics of each video block. The coding technique is adapted dynamically according to block size, content type, and prediction requirements, allowing the system to optimize performance for different scenarios without requiring all methods to be applied uniformly.
2Measurement precision
If Matrix-based Intra Prediction (MIP) with matrix vector multiplication is used, then prediction accuracy is improved, but computational complexity and processing time increase
Solution Approach 1:
The system applies MIP selectively to specific blocks where it provides the most benefit, rather than uniformly to all blocks. By using partial action (applying MIP only where needed) and combining it with faster prediction methods for other blocks, the system achieves high prediction accuracy for critical regions while maintaining acceptable overall processing time.
3Manufacturing precision
If Intra Subblock Partitioning (ISP) is applied to partition blocks into multiple sub-partitions, then coding precision is improved, but device complexity and processing overhead increase
Solution Approach 1:
ISP divides video blocks into multiple sub-partitions (e.g., 2×2 or 4 sub-blocks) that can be independently coded. This segmentation improves coding precision by allowing different prediction modes to be applied to different sub-partitions based on local characteristics, while the modular structure helps manage complexity through systematic processing of each sub-partition.
Solution Approach 2:
Each sub-partition created by ISP can have its own prediction mode and coding parameters optimized for local characteristics. This local quality approach allows the system to achieve high coding precision by adapting to local variations in video content, while the regular partitioning pattern provides a structured framework that manages processing complexity.
4Measurement precision
If Affine Linear Weighted Intra Prediction (ALWIP) is used for video blocks, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
ALWIP uses dynamic weighting factors that are adapted to each block's characteristics, allowing the prediction to be optimized for local content. The system dynamically adjusts the affine transformation parameters based on block size and content type, achieving high prediction accuracy while managing complexity through adaptive parameter selection rather than fixed complex computations.
Data Source
AI summary
A video processing method includes determining, for a conversion between a block of a video and a bitstream representation of the video, a manner in which information for a Matrix-based Intra Prediction (MIP) coding technique is coded in the bitstream representation. A prediction block of the block is determined using the MIP coding technique based on performing a matrix vector multiplication operation on previously coded samples of the video. The method also includes performing the conversion based on the determining.