Deblocking Filter Control for JCCR and ACT Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In video encoding and decoding processes, particularly in HEVC and VVC, the deblocking filter control based on non-zero transform coefficients fails to appropriately manage boundary filtering strength when adaptive color transform (ACT) or joint coding of chroma residual (JCCR) is applied, leading to image quality deterioration.
Innovation Solution
A deblocking filter device and decoding device that control boundary filtering strength based on whether adjacent blocks are encoded using ACT or JCCR, ensuring appropriate filtering even when these encoding tools are applied.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If deblocking filter control is based on non-zero transform coefficients, then filtering is applied when energy is distributed, but image quality deteriorates when ACT or JCCR is applied
Solution Approach 1:
The patent changes the control parameter for deblocking filter strength from transform coefficient-based criteria to prediction residual-based criteria. When ACT or JCCR is applied, the system evaluates the absolute values of prediction residuals at block boundaries to determine filtering strength, rather than relying on transform coefficient non-zero patterns that become unreliable under these encoding modes.
Solution Approach 2:
The patent introduces a feedback mechanism where the deblocking filter controller continuously evaluates the actual prediction residual values at block boundaries and adjusts filtering strength accordingly. This feedback loop ensures that filtering is applied appropriately based on actual boundary discontinuity rather than indirect proxies, preventing both over-filtering and under-filtering.
2Productivity
If JCCR is applied to generate joint prediction residual, then encoding efficiency improves, but boundary filtering control becomes inaccurate
Solution Approach 1:
The patent uses the prediction residual as an intermediary measure to bridge the gap between JCCR encoding and deblocking filter control. Instead of trying to control filtering based on the joint prediction residual generation process itself, the system measures the actual effect on block boundaries through prediction residual evaluation, providing accurate filtering control without interfering with JCCR encoding efficiency.
3Manufacturing precision
If ACT is applied for color space transform, then color representation improves, but deblocking filter control based on transform coefficients fails
Solution Approach 1:
The patent segments the deblocking filter control process into two independent parts: (1) the encoding process (including ACT) that operates on transform coefficients, and (2) the filtering control process that operates on prediction residuals. This segmentation allows ACT to improve color representation while the filtering control uses a different data source (prediction residuals) that is not affected by the color space transformation.
Data Source
AI summary
A deblocking filter device according to a first feature includes: a deblocking filter configured to perform a filter process on a boundary between a first reconstructed block and a second reconstructed block adjacent to the first reconstructed block; and a filter controller configured to control boundary filtering strength of the deblocking filter based on whether or not at least one of the first reconstructed block and the second reconstructed block is encoded using JCCR (Joint coding of chroma residual) in which one joint prediction residual is generated from prediction residuals of a Cb chrominance component and a Cr chrominance component.