Adaptive De-blocking Filter for MPEG Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reducing blocking artifacts in MPEG-coded video are either too complex for real-time applications or effective only at low bit rates, failing to provide satisfactory results at medium to high bit rates, which is a limitation in modern video compression standards.
Innovation Solution
A highly adaptive de-blocking filter that determines a filtering range and region mode based on local activity and quantization parameters, using symmetric or asymmetric filters to reduce blocking artifacts in both frame-coded and field-coded pictures, with adaptive threshold determination and interpolation for interlaced video.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional de-blocking filters are applied to reduce blocking artifacts, then blocking artifacts are reduced, but computational complexity increases significantly
Solution Approach 1:
The patent divides the filtering process into multiple stages: first applying a strong low-pass filter to reduce blocking artifacts, then applying a weak filter only when necessary. This segmentation of the filtering operation reduces computational complexity while maintaining effectiveness in reducing blocking artifacts.
Solution Approach 2:
The patent applies different filtering strengths to different regions of the image based on local characteristics. Strong filtering is applied to regions with significant blocking artifacts, while weak or no filtering is applied to regions where it would degrade quality. This local adaptation reduces overall computational complexity while maintaining artifact reduction effectiveness.
2Object-affected harmful factors
If strong filtering is applied to reduce blocking artifacts at low bit rates, then blocking artifacts are reduced, but sharpness is degraded
Solution Approach 1:
The patent dynamically adjusts the filtering strength based on the bit rate and local image characteristics. At low bit rates, stronger filtering is applied to reduce blocking artifacts. At medium to high bit rates, weaker or no filtering is applied to preserve sharpness. This dynamic adaptation resolves the contradiction between artifact reduction and sharpness preservation.
Solution Approach 2:
The patent changes the filtering parameters (strength, type) based on operating conditions. The system switches between strong low-pass filtering and weak filtering based on bit rate and local activity, optimizing the balance between blocking artifact reduction and sharpness preservation for each condition.
3Adaptability or versatility
If adaptive filtering is applied to handle different bit rates, then quality across bit rates is improved, but device complexity increases
Solution Approach 1:
The patent segments the adaptive filtering into simple conditional operations: check bit rate and local characteristics, then apply appropriate filter strength. This segmentation into basic decision points and simple filter applications achieves adaptability without excessive complexity.
Solution Approach 2:
The patent applies filtering only partially - using strong filtering only when necessary (low bit rates, smooth regions) and weak filtering only when needed (medium bit rates, boundary regions). This partial application of adaptive filtering achieves versatility across bit rates while keeping the overall system complexity manageable.
Data Source
AI summary
A post processing de-blocking filter includes a threshold determination unit for adaptively determining a plurality of threshold values according to at least differences in quantization parameters QPs of a plurality of adjacent blocks in a received video stream and to a user defined offset (UDO) allowing the threshold levels to be adjusted according to the UDO value; an interpolation unit for performing an interpolation operation to estimate pixel values in an interlaced field if the video stream comprises interlaced video; and a de-blocking filtering unit for determining a filtering range specifying a maximum number of pixels to filter around a block boundary between the adjacent blocks, determining a region mode according to local activity around the block boundary, selecting one of a plurality of at least three filters, and filtering a plurality of pixels around the block boundary according to the filtering range, the region mode, and the selected filter.