Joint Sub-pixel Interpolation Filter for HEVC Temporal Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-efficiency video coding (HEVC) faces challenges in accurately interpolating sub-pixel values for temporal motion prediction, as conventional interpolation filters do not effectively account for sub-pixel offsets of multiple reference blocks, leading to signal distortion and noise issues.
Innovation Solution
The method involves determining a set of interpolation filters based on sub-pel offsets for two reference blocks, allowing for joint sub-pixel interpolation that considers both reference blocks' movements, thereby improving the accuracy of sub-pel pixel value interpolation in temporal prediction processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional interpolation filters are used for temporal motion prediction, then the filtering process is simple and fast, but the accuracy of sub-pixel interpolation is poor and signal distortion occurs
Solution Approach 1:
The patent applies dynamics by making the interpolation filter adaptable to different sub-pixel offset conditions. Instead of using a fixed conventional filter, the system dynamically selects or adjusts filters based on the specific sub-pixel offsets of reference blocks, allowing the filter characteristics to change according to the motion prediction requirements. This resolves the contradiction by enabling high accuracy when needed while maintaining simplicity through selective application.
Solution Approach 2:
The patent changes the parameters of the interpolation filter based on sub-pixel offset values. By adjusting filter coefficients and tap lengths according to the specific sub-pixel positions of reference blocks, the system optimizes interpolation accuracy for different motion scenarios. This parameter adaptation allows the filter to maintain high precision across varying conditions without requiring a completely complex system architecture.
2Measurement precision
If separate interpolation filters are used for each reference block, then the interpolation accuracy for each block is optimized, but the overall processing complexity increases
Solution Approach 1:
The patent merges the interpolation filtering operations for multiple reference blocks into a unified joint filtering process. Instead of independently applying separate filters to each reference block and then combining results, the system performs a single joint interpolation that simultaneously accounts for sub-pixel offsets of both reference blocks. This merging approach maintains optimization for each block while reducing overall processing complexity through operational consolidation.
Solution Approach 2:
The patent creates a universal joint interpolation filter that serves multiple reference blocks simultaneously. This multi-functional filter design allows a single filtering operation to handle interpolation for both reference blocks in bi-prediction scenarios, eliminating the need for separate dedicated filters for each block. The universal filter maintains accuracy for each block while simplifying the overall system architecture.
3Measurement precision
If HEVC uses specific interpolation filters for each reference block based on sub-pel position, then sub-pel accuracy is improved, but signal distortion and noise issues arise due to inconsistent filter application
Solution Approach 1:
The patent applies homogeneity by ensuring consistent filter application across multiple reference blocks in bi-prediction. The joint interpolation filter uses uniform filtering operations for both reference blocks, eliminating the inconsistency that causes signal distortion and noise. This homogeneous approach maintains sub-pel accuracy while ensuring that both reference blocks are processed with the same filtering characteristics, preventing harmful artifacts.
Solution Approach 2:
The patent introduces a joint interpolation filter as an intermediary that mediates between multiple reference blocks with different sub-pixel offsets. This intermediary filter harmonizes the interpolation process by applying coordinated filtering that accounts for both reference blocks' positions simultaneously. The mediator filter prevents the signal distortion and noise that arise from inconsistent separate filtering by providing a unified interpolation approach.
Data Source
AI summary
A method determines a plurality of interpolation filters for use in interpolating sub-pel values for a temporal prediction process of video content. The plurality of interpolation filters are designed based on sub-pel offsets for two reference blocks. The method determines a first sub-pel offset for a first reference block for the unit of video content and determines a second sub-pel offset for a second reference block for a unit of video content. A set of interpolation filters is determined to interpolate a set of sub-pel pixel values for use in the temporal prediction process for the unit of video content. The set of interpolation filters is designed for the first sub-pel offset and the second sub-pel offset and is used to interpolate a first sub-pel pixel value for the first reference block and a second sub-pel value for the second reference block.


