Deblocking Filtering for IPCM Block Boundaries in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing H.264 video compression standard lacks appropriate filtering methods for the boundaries between Intra Pulse Code Modulation (IPCM) and non-IPCM blocks, leading to deterioration in image quality due to uniform filter strength application, which fails to differentiate and control filtering effectively across block types.
Innovation Solution
A method for determining a filter strength by setting a second quantization parameter for the IPCM block based on the non-IPCM block's quantization parameter, allowing for differential filtering between IPCM and non-IPCM blocks, thereby enabling more appropriate deblocking filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If uniform filter strength is applied to all blocks including IPCM blocks, then the filtering process is simple and consistent, but image quality deteriorates due to inappropriate filtering on IPCM block boundaries
Solution Approach 1:
The patent applies different filter strengths to different block types by introducing a flag that identifies IPCM blocks. When this flag is set, the filter strength is reduced or filtering is skipped for IPCM blocks, while normal filtering is applied to non-IPCM blocks. This local differentiation resolves the contradiction by making the filtering process adaptive to block characteristics rather than uniformly applied.
Solution Approach 2:
The patent changes the filter strength parameter based on block type. By introducing a flag that modifies the filter strength calculation, the system dynamically adjusts the filtering intensity: strong filtering for non-IPCM blocks and weak or no filtering for IPCM blocks. This parameter change enables appropriate filtering for each block type while maintaining a relatively simple overall process.
2Object-affected harmful factors
If deblocking filtering is performed on IPCM block boundaries, then block artifacts are reduced, but original pixel values are distorted due to unnecessary filtering on uncompressed blocks
Solution Approach 1:
The patent introduces a flag to identify IPCM blocks and applies filtering selectively based on this identification. For IPCM blocks, the filter strength is reduced or filtering is skipped entirely, preserving the original pixel values. For non-IPCM blocks, normal deblocking filtering is applied to reduce artifacts. This local quality approach ensures that filtering is applied only where necessary.
Solution Approach 2:
The patent converts the potential harm of over-filtering IPCM blocks into a benefit by using the uncompressed nature of IPCM blocks as an indicator for reduced filtering. The fact that IPCM blocks are already uncompressed and high-quality is transformed into a criterion for applying lighter filtering, thereby preserving their original quality while still maintaining block boundary consistency.
3Manufacturing precision
If filter strength is determined based on quantization parameters of both adjacent blocks, then filtering is optimized for compressed blocks, but IPCM blocks receive inappropriate strong filtering due to their unique characteristics
Solution Approach 1:
The patent modifies the filter strength determination process by introducing a flag that identifies IPCM blocks. When this flag is set for either adjacent block, the filter strength is reduced or filtering is skipped for the IPCM block, regardless of the quantization parameters. This local quality differentiation enables the system to adapt to different block types while still using quantization parameters for non-IPCM blocks.
Solution Approach 2:
The patent makes the filtering process dynamic by introducing a flag-based mechanism that adjusts filter strength based on block type identification. The filter strength is no longer statically determined solely by quantization parameters but dynamically adjusted according to whether blocks are IPCM or non-IPCM type. This dynamic adaptation allows the system to optimize filtering for each block type's characteristics.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A filtering method according to the present invention is for performing deblocking filtering on the boundary between an IPCM block and a non-IPCM block adjacent to each other in an image and including: determining a first quantization parameter for the non-IPCM block (540.1); determining a second quantization parameter for the IPCM block, using the first quantization parameter (5402); determining a filter strength for the boundary, using the first quantization parameter and the second quantization parameter (S403); and performing the deblocking filtering on the boundary using the determined filter strength (S404).