Nonlinear Adaptive Loop Filter for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently reducing redundancy and improving coding efficiency, particularly in handling intra prediction and motion compensation with varying prediction directions and bit depths.
Innovation Solution
The implementation of a nonlinear adaptive loop filter with a clipping function, where the clipping value is determined by a multiplication of functions dependent on the bit depth and clipping index, allowing for adaptive filtering based on the bit depth and filtering strength.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional linear loop filtering is used in video coding, then the implementation is simple, but the coding efficiency and video quality are insufficient
Solution Approach 1:
The patent applies parameter changes by transitioning from linear filtering coefficients to non-linear clipping values that adapt to different bit depths and prediction directions. The clipping value is calculated as a function of bit depth B and prediction direction, allowing the filter to dynamically adjust its parameters based on the specific coding conditions, thereby improving video quality while maintaining manageable complexity through a systematic parameter adaptation approach.
Solution Approach 2:
The patent implements dynamics by making the loop filter adaptive rather than static. The clipping values change dynamically based on the bit depth and prediction direction parameters, allowing the filtering process to automatically adapt to different coding scenarios. This dynamic adaptation enables the filter to optimize performance for each specific case without requiring manual configuration or complex manual tuning.
2Productivity
If non-linear adaptive loop filtering with bit depth dependent clipping is implemented, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the filtering process into distinct stages: first determining the bit depth parameter, then calculating the clipping value based on that bit depth and the prediction direction. This segmented approach allows each stage to be optimized independently, with the clipping value calculation being a simple function of already-available parameters, thereby improving coding efficiency without proportionally increasing computational complexity.
Solution Approach 2:
The patent implements self-service by making the filter automatically adapt to different bit depths and prediction directions without requiring external control or manual intervention. The clipping values are self-determined based on the intrinsic parameters of the coding process (bit depth and prediction direction), allowing the system to optimize its own performance automatically for each coding scenario.
3Adaptability or versatility
If fixed clipping values are used in loop filtering, then the implementation is straightforward, but adaptability to different bit depths and prediction directions is limited
Solution Approach 1:
The patent applies universality by creating a clipping value calculation mechanism that works across all bit depths and prediction directions through a single unified function. Rather than requiring separate clipping values for each combination of bit depth and prediction direction, the patent uses a universal formula that adapts to any input parameters, thereby achieving high adaptability without increasing configuration complexity.
Solution Approach 2:
The patent implements parameter changes by making the clipping values dependent on the bit depth and prediction direction parameters. Instead of using fixed clipping values, the system dynamically adjusts the clipping parameters based on the specific coding conditions, allowing the same filter structure to adapt to various bit depths (e.g., 8-bit, 10-bit, 12-bit) and prediction directions without requiring separate filter configurations.
Data Source
AI summary
Aspects of the disclosure provide a method and an apparatus including processing circuitry for video decoding. The processing circuitry can decode coded information of a coding block in a picture of a coded video sequence. The coded information indicates a clipping index m. The clipping index m indicates a clipping value for a filter that is applied to the coding block. The processing circuitry can determine the clipping value associated with the clipping index m. The clipping value can be dependent on a bit depth B and the clipping index m. The clipping value for each value of the clipping index m can be independent from N. N is a total number of allowed clipping values that include the clipping value. The processing circuitry generates a filtered coding block by applying the filter corresponding to the clipping value to the coding block.