Video Coding Frame Boundary Artifact Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional video coding methods, such as AVC and HEVC, often generate image artifacts due to interpolation of image content from edges of reference frames, particularly in skip mode coding, leading to 'smear' artifacts in regions with low motion content.
Innovation Solution
The proposed solution involves estimating the extent to which a prediction block exceeds the edge of a reference frame and partitioning the pixel block into sub-blocks when the excess exceeds a threshold, allowing for new motion prediction and coding mode selection for each sub-block, and applying a bias against skip mode coding to reduce distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If skip mode coding is used for low motion content regions, then coding efficiency is improved, but image artifacts (smear artifacts) are generated due to interpolation from reference frame edges
Solution Approach 1:
The current frame is divided into multiple blocks, and each block is independently evaluated for skip mode applicability. Blocks near reference frame edges are identified and excluded from skip mode coding, while interior blocks can still use skip mode for efficient compression.
Solution Approach 2:
Different coding modes are applied to different regions of the frame based on local characteristics. Edge regions use modes that prevent artifact generation, while interior low-motion regions use skip mode for efficiency, creating locally optimized coding quality.
2Measurement precision
If motion prediction is performed using reference blocks from reference frames, then prediction accuracy is improved, but frame boundary artifacts occur when prediction blocks extend beyond reference frame edges
Solution Approach 1:
Before performing motion prediction, the method预先 determines whether the prediction block would extend beyond reference frame edges. This preliminary check prevents artifact generation by avoiding problematic prediction operations before they occur.
Solution Approach 2:
When a prediction block would extend beyond the reference frame edge, an intermediary reference block is constructed by replicating available reference pixels within the frame boundaries, rather than directly accessing out-of-bound memory locations. This intermediary structure prevents artifacts while maintaining prediction functionality.
3Manufacturing precision
If the encoder maximizes decoded visual quality under target bitrate constraint, then visual quality is improved, but device complexity increases due to multiple coding mode evaluations
Solution Approach 1:
The frame is segmented into edge-adjacent blocks and interior blocks, with different evaluation rules applied to each segment. This reduces the overall complexity by applying simpler rules to edge blocks while maintaining quality optimization for interior blocks.
Solution Approach 2:
The coding decision parameters are changed based on block location. For edge-adjacent blocks, the skip mode parameter is disabled or heavily penalized, while interior blocks maintain full parameter flexibility for quality optimization.
Data Source
AI summary
Techniques for encoding video with reduced frame-boundary artifacts are presented. The techniques include a video encoding method, where, when skip mode is selected as a motion prediction coding mode for a pixel block predicted from a reference block of a reference frame, estimating an amount of the reference block that extends beyond the edge of the reference frame. If the amount beyond the edge does not exceed a threshold, the prediction may be coded with skip mode. If the amount beyond the edge exceeds a threshold, the pixel block may be partitioned, and the partitioned blocks may be coded with motion prediction.