Adaptive Loop Filter with Non-Sample Taps for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding systems, such as VVC, face challenges in effectively utilizing Adaptive Loop Filters (ALF) due to limitations in filter adaptability and efficiency, particularly in handling chroma components and cross-component filtering, leading to suboptimal video quality.
Innovation Solution
Introduce non-sample-value taps in the ALF reconstruction process, utilizing positional and target information unrelated to sample values, such as positional functions and CU-coded information, to enhance filter adaptability and improve video quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional ALF filter taps are used (based on sample values), then the filter can process chroma components, but the adaptability and efficiency are limited
Solution Approach 1:
The patent introduces non-sample-value taps that operate in a different dimension from traditional sample-value taps. These taps use positional information and target information (such as block type, prediction mode, motion vector) to generate filter coefficients, adding a new dimension to the filtering process that enables better adaptability without sacrificing efficiency
Solution Approach 2:
The non-sample-value taps serve multiple functions simultaneously: they provide adaptability across different block types and prediction modes while also maintaining filtering efficiency. The same tap mechanism works for both luma and chroma components, and can be applied across different coding scenarios, making the filter universally applicable
2Adaptability or versatility
If non-sample-value taps are introduced, then filter adaptability is enhanced, but device complexity increases
Solution Approach 1:
The patent merges the non-sample-value tap generation with the existing ALF filter structure. The positional and target information processing is integrated into the same functional block that handles sample-value taps, rather than creating a separate complex system. This merging approach enhances adaptability while minimizing the increase in overall device complexity
3Manufacturing precision
If more filter taps are used, then video quality improves, but processing time increases
Solution Approach 1:
The patent changes the parameters used for tap generation from sample values to non-sample values (positional and target information). This parameter change allows the filter to achieve higher video quality with the same or fewer taps, because the non-sample values provide more meaningful adaptability cues that reduce the need for additional tap combinations, thereby decreasing processing time
Data Source
AI summary
A method and apparatus for video coding using ALF with non-sample-value filter taps. According to the method, reconstructed pixels are received, wherein the reconstructed pixels comprise a current block. A current filtered output from an ALF for a current sample in the current block is derived, wherein the ALF comprises at least one non-sample-value term derived using target information unrelated to sample values of the current block and the target information is derived at an encoder side, or derived or received at a decoder side. Filtered-reconstructed pixels are provided, wherein the filtered-reconstructed pixels comprise the current filtered output.