Video Deblocking Filter Strength from Motion Vector Differences
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently applying deblocking filters, particularly in scenarios involving intra block copy (IBC) coding, residue differential pulse-code modulation (RDPCM), palette coding, decoder-side motion refinement, and other advanced coding tools, leading to suboptimal video quality and bandwidth utilization.
Innovation Solution
The proposed method determines boundary strength for deblocking filters based on various criteria such as motion vector differences, intra block copy coding, and other advanced coding tools, selectively applying the filter to enhance video quality by optimizing filter strength and application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If deblocking filters are applied to all block boundaries, then block artifacts are reduced, but processing complexity and bandwidth increase
Solution Approach 1:
The patent applies deblocking filters selectively based on local block characteristics rather than uniformly across all boundaries. Boundary strength values (BS=0,1,2) are assigned differently depending on whether blocks are intra-coded, inter-coded, or IBC-coded, and whether they use RDPCM or palette coding. This localized approach reduces processing complexity while maintaining quality where needed.
Solution Approach 2:
The patent changes the parameter of boundary strength assignment based on coding mode. Specifically, it modifies how boundary strength is determined for IBC-coded blocks with RDPCM or palette coding (setting BS=0 for certain cases) versus traditional coding modes, thereby optimizing the balance between artifact reduction and processing efficiency for different local conditions.
2Manufacturing precision
If deblocking filters are applied to all block boundaries, then video quality is improved, but bandwidth consumption increases
Solution Approach 1:
The patent applies deblocking filters only where necessary based on local block characteristics. By identifying IBC-coded blocks with RDPCM or palette coding and assigning them boundary strength=0, the filter is skipped for these regions, reducing bandwidth consumption while maintaining quality in regions where filtering is actually needed.
Solution Approach 2:
The patent extracts and identifies specific block types (IBC-coded blocks with RDPCM or palette coding) from the overall video data and applies a different filtering strategy to them. By separating these special cases and excluding them from standard deblocking processing, the patent reduces unnecessary bandwidth consumption while preserving video quality through targeted filtering of other block types.
3Measurement precision
If boundary strength is determined using motion vector difference, then filtering accuracy is improved, but calculation complexity increases
Solution Approach 1:
Instead of calculating motion vector differences for all blocks to determine boundary strength, the patent inverts the approach by first identifying special block types (IBC-coded with RDPCM or palette coding) and directly assigning them boundary strength=0 without motion vector calculation. This eliminates unnecessary calculations while maintaining accurate boundary strength determination for other block types.
Solution Approach 2:
The patent extracts IBC-coded blocks with RDPCM or palette coding from the general block population and applies a simplified boundary strength determination method to them. By separating these blocks and using a direct assignment rule (BS=0) instead of motion vector difference calculation, the patent reduces calculation complexity while preserving filtering accuracy for blocks where it matters.
Data Source
AI summary
A method for visual media processing is disclosed. The method includes determining whether a pair of adjacent blocks of visual media data are both intra block copy (IBC) coded and selectively applying, based on the determination, a deblocking filter (DB) process by identifying a boundary at a vertical edge and/or a horizontal edge of the pair of adjacent blocks. The method further includes calculating a boundary strength of a filter, deciding whether to turn on or off the filter, and selecting a strength of the filter in case the filter is turned on. The boundary strength of the filter is dependent on a motion vector difference between motion vectors associated with the pair of adjacent blocks.