Video Block Transform Kernel Mapping for Interpolation Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding and decoding technologies face challenges in improving prediction accuracy and coding efficiency due to inefficiencies in determining transform kernels for video blocks, leading to suboptimal decoding accuracy and increased codeword usage.
Innovation Solution
The proposed method involves determining a reference region and interpolation filter for a current block, using these to predict a block and determine a transform kernel that aligns with the block's characteristics, thereby improving decoding accuracy and reducing the need for separate codeword indication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction modes are used to determine transform kernels, then device complexity is reduced, but decoding accuracy deteriorates
Solution Approach 1:
The patent changes the parameter used for transform kernel determination from traditional prediction modes to interpolation filter types. By establishing a correspondence relationship between interpolation filter types and transform kernels, the system achieves more accurate decoding while maintaining manageable complexity through parameter mapping rather than complex computational processes.
Solution Approach 2:
The patent uses the interpolation filter type as a copy or proxy indicator to determine the transform kernel. Instead of directly computing the optimal transform kernel through complex analysis, the system copies the filter type information and maps it to the corresponding transform kernel, simplifying the determination process while improving accuracy.
2Adaptability or versatility
If separate codeword indication is used for transform kernels, then adaptability is improved, but codeword usage increases
Solution Approach 1:
The patent merges the transform kernel indication with the existing interpolation filter type information. By combining these two previously separate elements into a single correspondence relationship, the system eliminates the need for separate codeword indication while maintaining full adaptability of transform kernel selection across different block types and characteristics.
Solution Approach 2:
The interpolation filter type serves multiple functions simultaneously: it characterizes the block's frequency content properties and also indicates the appropriate transform kernel to use. This multi-functionality eliminates the need for dedicated codewords for transform kernel selection, reducing overall codeword usage while preserving adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present application provides a video coding method and apparatus, a video decoding method and apparatus, devices, a system, and a storage medium. The method comprises: when predicting the current block, determining a reference area and an interpolation filter of the current block, and determining a prediction block of the current block on the basis of the reference area and the interpolation filter; determining a prediction mode corresponding to the prediction block, so as to determine, on the basis of the prediction mode, a transformation kernel corresponding to the current block; and using the transformation kernel to perform inverse transformation on a transformation coefficient of the current block to obtain a residual block of the current block, and obtaining a reconstruction value of the current block on the basis of the residual block and the prediction block of the current block. According to the present application, if the current block is predicted by using an interpolation filtering prediction method, a transformation kernel corresponding to the current block is determined by determining a conventional prediction mode corresponding to the prediction block, thereby improving the determination accuracy of the transformation kernel, and improving the video coding and decoding effects of the current block.