Adaptive Pixel Classification for Video In-Loop Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video coding standards, such as HEVC, face limitations in accurately correcting errors in reconstructed images due to the restricted nature of edge and band offsets, which are insufficient for handling the diverse features of modern images.
Innovation Solution
An adaptive pixel classification method is introduced for in-loop filtering, where reconstructed samples are classified using either absolute or relative standards, and an offset value is added based on the classification results, allowing for more precise error correction and feature-specific offset application.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional edge offset and band offset are used for sample adaptive offset, then the filtering process is simple, but the error correction capability is limited due to restricted classification standards
Solution Approach 1:
The reconstructed sample is divided into multiple categories using multiple classification standards (absolute classification and relative classification). Each category receives a specific offset value tailored to its characteristics, enabling precise error correction for different image features such as edges, bands, and flat regions without overwhelming complexity
Solution Approach 2:
Different offset values are applied to different categories of reconstructed samples based on their local characteristics. The absolute classification standard identifies samples by brightness value bands, while the relative classification standard identifies samples by edge and gradient features, allowing each local region to receive appropriate correction
2Adaptability or versatility
If multiple classification standards are applied for adaptive pixel classification, then the adaptability to diverse image features is improved, but the computational complexity increases
Solution Approach 1:
The classification process is segmented into two independent standards: absolute classification (based on brightness value bands) and relative classification (based on edge and gradient information). This segmentation allows the system to handle diverse image features adaptively while maintaining manageable computational complexity through modular processing
Solution Approach 2:
The filtering system dynamically selects and applies appropriate classification standards based on the characteristics of each reconstructed sample. The dual classification approach enables the system to adapt to varying image features (edges, bands, flat regions) while the offset values are dynamically adjusted according to classification results
Data Source
AI summary
The in-loop filtering method performed by a video decoding apparatus includes: classifying reconstructed samples according to an absolute classification standard or relative classification standard; acquiring offset data on the basis of results of classifying the reconstructed samples; adding an offset value to the reconstructed samples by referencing the acquired offset data; and outputting the offset value-added reconstructed samples. Accordingly, W errors in the reconstructed image can be corrected.