Adaptive Filtering Reference Pixels Block Size Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video compression techniques degrade the accuracy of intra-prediction by filtering reference pixels without considering the characteristics of neighboring blocks, leading to inefficient image encoding and decoding.
Innovation Solution
An adaptive filtering method that adjusts the filtering of reference pixels based on the relative size of neighboring blocks with respect to the current block, applying stronger or weaker filters depending on block sizes and presence of edges, to enhance prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional filtering is applied to reference pixels without considering neighboring block characteristics, then filtering process is simple, but prediction accuracy is degraded
Solution Approach 1:
The patent applies different filtering strengths (strong, medium, weak, or no filtering) to reference pixels based on the local characteristics of neighboring blocks. Specifically, when the neighboring block size is larger than the current block, strong filtering is applied; when sizes are equal, medium filtering is applied; when the neighboring block is smaller, weak or no filtering is applied. This localized adaptation of filtering intensity resolves the contradiction by tailoring the filtering process to local block characteristics, improving prediction accuracy without uniformly increasing complexity across all blocks.
Solution Approach 2:
The filtering process is made dynamic by adaptively selecting filtering strength based on the size relationship between neighboring blocks and current blocks. The system dynamically adjusts the filtering operation rather than applying a fixed filtering process, allowing the filtering behavior to change according to the specific block configuration being processed.
2Measurement precision
If adaptive filtering based on block size relationship is applied, then prediction accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent applies adaptive filtering selectively based on local block size relationships rather than uniformly processing all reference pixels with the same filtering strength. This allows the system to focus computational resources on blocks where adaptive filtering provides the most benefit (larger neighboring blocks) while reducing or eliminating filtering for blocks where it provides minimal improvement (smaller neighboring blocks), thereby improving overall encoding efficiency.
Solution Approach 2:
The filtering operation parameters (filtering strength, filter type) are changed adaptively based on the block size relationship. The system modifies the filtering parameters according to the specific configuration of neighboring blocks, applying stronger filtering when block sizes indicate higher correlation and weaker or no filtering when block sizes suggest lower correlation, thus optimizing the balance between accuracy and efficiency.
3Quantity of substance
If reference pixels from quantized neighboring blocks are used for intra-prediction, then data availability is improved, but quantization errors are introduced
Solution Approach 1:
The patent converts the harmful effect of quantization errors into a beneficial filtering opportunity. By recognizing that quantized reference pixels contain errors, the system applies adaptive filtering to remove or reduce these quantization artifacts. The filtering process transforms the problematic quantized data into improved reference pixels with reduced errors, turning the initial disadvantage into an advantage for prediction accuracy.
Solution Approach 2:
The system changes the parameters of reference pixels by applying filtering operations that modify pixel values to reduce quantization errors. The filtering process adjusts the reference pixel parameters based on local block characteristics, transforming the degraded quantized data into corrected reference data suitable for accurate intra-prediction.
Data Source
AI summary
Disclosed is a method for encoding and decoding an image by applying different filtering strength to reference pixels based on a size relationship between a current block and each reference block such that a prediction block for the current block is constructed using reference pixels which are filtered adaptively according to their properties.