Adaptive High-Pass Filter for Video Motion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems face inefficiencies in motion compensated prediction due to missing or distorted high-frequency information caused by noise and quantization errors, particularly affecting the luma and chroma components, which reduces coding efficiency and quality.
Innovation Solution
The implementation of adaptive filtering techniques, including high-pass filtering for the luma component and cross-plane filtering using high-frequency information from the luma component to enhance the chroma signals, optimizes filter coefficients for minimal energy residual and applies these filters during motion compensation to improve prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion compensated prediction is used in video coding, then compression efficiency is improved, but high-frequency information is lost or distorted due to noise and quantization errors
Solution Approach 1:
A high-pass filter is introduced as an intermediary component in the motion compensation pipeline. This filter processes the motion compensated prediction signal to restore high-frequency information that was lost during compression, effectively mediating between the compressed reference picture and the current picture being encoded
Solution Approach 2:
The high-pass filter is applied preliminarily to the motion compensated prediction signal before it is used for predicting the current picture. This preliminary restoration of high-frequency content ensures that subsequent encoding operations work with enhanced signal quality, preventing further degradation
2Device complexity
If traditional motion compensation is applied, then coding complexity is reduced, but prediction quality deteriorates due to missing high-frequency details
Solution Approach 1:
The high-pass filter selectively restores high-frequency information only in regions where motion compensation is applied, rather than processing the entire picture uniformly. This localized approach maintains prediction quality where needed while avoiding unnecessary complexity in static regions
3Manufacturing precision
If high-pass filtering is applied to restore high-frequency information, then prediction quality is improved, but residual energy increases
Solution Approach 1:
The system employs feedback mechanisms where the filtered prediction signal is compared with the original picture, and the resulting residuals are analyzed to adaptively adjust filtering parameters. This feedback loop ensures that filtering strength is optimized to minimize residual energy while maintaining prediction quality
Data Source
AI summary
Video coding systems and methods are described using an enhanced motion compensated picture. In exemplary embodiments, an enhanced motion compensated picture is generated by applying at least one high-pass filter to the motion compensated picture and adding the output of the filter to the motion compensated picture. Coefficients of the high-pass filter are selected by comparing the enhanced motion compensated picture to an original picture. The selected coefficients may be quantized and entropy coded into a bit stream. The high-pass filter may be a cross-plane filter, in which a luma component, which may be an enhanced luma component, is high-pass filtered and the output of the filter is added to at least one of the chroma components to generate an enhanced chroma component.