Bilateral and CCSAO In-Loop Filtering for Chroma Ringing Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies struggle to effectively remove ringing artifacts, particularly in chroma components, despite the use of in-loop filters like deblocking, sample adaptive offset (SAO), and adaptive loop filter (ALF).
Innovation Solution
The integration of a bilateral filter with a cross-component sample adaptive offset (CCSAO) filter, where the bilateral filter can be applied before, after, or in parallel with CCSAO, and their offsets can be used to generate offsets for each other, along with conditional application based on thresholds and statistical information, to enhance compression efficiency and reduce artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If in-loop filters (deblocking, SAO, ALF) are applied to remove artifacts, then video quality is improved, but ringing artifacts in chroma components persist
Solution Approach 1:
The patent combines bilateral filtering with CCSAO filtering into a unified in-loop filtering process. The bilateral filter and CCSAO filter work together synergistically, where the bilateral filter reduces ringing artifacts through its distance-weighted and intensity-weighted filtering mechanism, while CCSAO provides cross-component sample adaptive offset correction. This merged approach effectively removes ringing artifacts in chroma components that traditional filters alone cannot eliminate.
Solution Approach 2:
The patent employs a composite filtering approach by integrating two different filtering mechanisms (bilateral filter and CCSAO filter) into a single processing pipeline. The bilateral filter uses a composite weighting function that combines spatial distance weights and intensity difference weights, creating a composite filtering effect that adapts to local image characteristics and effectively suppresses ringing artifacts while preserving edges.
2Object-affected harmful factors
If bilateral filter is applied with varying filter weights based on distance and intensity differences, then ringing artifacts are reduced, but computational complexity increases
Solution Approach 1:
The bilateral filter applies different filter weights to different samples based on their local characteristics. Samples closer to the central sample and with similar intensity values receive higher weights, while distant or dissimilar samples receive lower weights. This local quality approach allows the filter to adapt to local image structures, effectively reducing ringing artifacts in smooth regions while preserving edges, without requiring uniform complex processing across the entire image.
Solution Approach 2:
The patent applies bilateral filtering selectively rather than uniformly across all samples. By using CCSAO to identify regions where filtering is most beneficial and applying bilateral filtering primarily in those regions, the system achieves effective artifact removal without the full computational cost of applying bilateral filtering to every sample in the image.
3Productivity
If CCSAO filter and bilateral filter are applied in sequence or parallel, then compression efficiency is improved, but processing time increases
Solution Approach 1:
The CCSAO filter is applied before the bilateral filter in the processing pipeline. The CCSAO filter performs cross-component sample adaptive offset correction first, preparing the data by correcting chroma samples using luma information. This preliminary action simplifies the subsequent bilateral filtering operation, as the data is already partially optimized, reducing the computational burden and processing time of the bilateral filter while maintaining compression efficiency.
Data Source
AI summary
A mechanism for processing video data is disclosed. The mechanism determines to apply a bilateral filter and a cross component sample adaptive offset (CCSAO) filter to samples in a current block of a current picture. The bilateral filter includes filter weights that vary based on a distance between surrounding samples and a central sample and differences in intensities of the surrounding samples and the central sample. A conversion is performed between a visual media data and a bitstream based on the bilateral filter and the CCSAO filter.


