Asymmetric In-Loop Filtering Across Virtual Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video codecs face challenges in performing in-loop filtering across virtual boundaries in multimedia systems, leading to potential leaks and mismatches between encoder and decoder due to incorrect decoding of non-refreshed areas.
Innovation Solution
Implementing asymmetric in-loop filters that allow filtering from one side of a virtual boundary to another while preventing the use of incorrect decoding information from the other side, using padding or predetermined values to maintain accurate filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If symmetric in-loop filtering is applied across virtual boundaries, then filtering consistency is improved, but decoding accuracy deteriorates due to incorrect decoding information from non-refreshed areas
Solution Approach 1:
The patent implements asymmetric in-loop filtering at virtual boundaries by applying different filtering operations to pixels on opposite sides of the boundary. Specifically, pixels in the refreshed area use filtering that prevents dependency on non-refreshed area data, while pixels in the non-refreshed area use filtering that compensates for the asymmetric treatment. This asymmetric approach resolves the contradiction by maintaining filtering consistency through coordinated asymmetric operations rather than identical symmetric operations.
Solution Approach 2:
The patent applies different filtering strategies to different local regions around the virtual boundary. Pixels adjacent to the boundary in the refreshed area receive different treatment than pixels in the non-refreshed area, with each region's filtering tailored to its specific decoding constraints. This local differentiation allows the system to maintain overall filtering consistency while preserving decoding accuracy in each local region.
2Manufacturing precision
If in-loop filtering uses coding information from both sides of virtual boundary, then filtering quality is improved, but reliability deteriorates due to leaks from incorrect decoding
Solution Approach 1:
The patent segments the filtering process into distinct operations for pixels on opposite sides of the virtual boundary. Pixels in the refreshed area undergo filtering that segments out dependencies on non-refreshed area data, while pixels in the non-refreshed area undergo separate filtering that compensates for this segmentation. This segmentation approach maintains filtering quality by ensuring each segment operates with reliable data while preserving overall filtering effectiveness.
Solution Approach 2:
The patent introduces an intermediary mechanism at the virtual boundary that mediates the flow of coding information between refreshed and non-refreshed areas. This intermediary prevents direct use of potentially incorrect decoding information from non-refreshed areas while still allowing filtered pixels to contribute to their respective areas' reconstruction. The intermediary maintains filtering quality by controlling information flow rather than completely blocking it.
3Measurement precision
If asymmetric filtering is applied at virtual boundaries, then decoding accuracy is improved, but device complexity increases due to different filtering operations
Solution Approach 1:
The patent applies asymmetric filtering selectively only to pixels adjacent to virtual boundaries, while interior pixels continue to receive standard symmetric filtering treatment. This partial application of asymmetric operations maintains decoding accuracy where it is most needed (at boundaries) while minimizing the increase in device complexity by avoiding unnecessary asymmetric processing in interior regions.
Data Source
AI summary
In accordance with example embodiments of the invention there is at least a method and an apparatus to perform: determining a virtual boundary that separates a picture, or a portion of the picture, into a first area and a second area; and determining to perform filtering of at least one pixel of the first area with coding information of the second area derived from the first area or with the coding information of the second area set to at least one value, when the coding information of the second area is to be used to perform the filtering of the at least one pixel of the first area, or determine to not perform the filtering of the at least one pixel of the first area, when the coding information of the second area is to be used to perform the filtering of the at least one pixel of the first area.


