Adaptive Intra-Prediction Filter for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies, such as the H. 264/AVC standard, face inefficiencies in intra-prediction processes, particularly when filters are applied before prediction, which can hinder coding performance.
Innovation Solution
An adaptive filter approach is implemented, where decisions to apply a first filter to reference pixel values and a second filter to prediction values are made based on information about surrounding blocks, prediction modes, block sizes, and coding types, allowing for optimized filtering during intra-prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a filter is applied to reference pixel values before intra-prediction, then prediction accuracy may be improved, but coding performance deteriorates
Solution Approach 1:
The patent applies filtering dynamically based on the prediction mode. For directional prediction modes (modes 1-8), filtering is applied to reference pixels to improve prediction accuracy. For non-directional modes (modes 0, 2), filtering is not applied to maintain coding efficiency. This dynamic approach resolves the contradiction by adapting the filtering operation to the specific prediction context.
Solution Approach 2:
The patent applies different filtering strategies to different parts of the prediction process. Specifically, filtering is applied selectively to certain reference pixels based on their position and the prediction mode being used. This local differentiation allows the system to improve prediction accuracy where beneficial while avoiding performance degradation where filtering would be harmful.
2Stability of the object's composition
If filtering is applied to all reference pixels, then prediction smoothness is improved, but encoding complexity increases
Solution Approach 1:
The patent applies filtering selectively to specific reference pixels based on the prediction mode. For example, in directional prediction modes, only certain reference pixels along the prediction direction are filtered, while others remain unfiltered. This localized filtering approach improves prediction smoothness in regions where it matters while minimizing the overall encoding complexity.
Solution Approach 2:
Instead of applying filtering to all reference pixels uniformly, the patent applies filtering only to the necessary subset of pixels required for accurate prediction in each mode. This partial action approach achieves sufficient prediction smoothness without the excessive complexity of universal filtering.
3Productivity
If adaptive filtering is applied based on surrounding block information, then coding efficiency is improved, but processing overhead increases
Solution Approach 1:
The patent changes the parameter of filtering application based on surrounding block characteristics. Specifically, the decision to filter is made based on the prediction mode of surrounding blocks and the current block's prediction mode. This parameter-based adaptation improves coding efficiency by matching filtering behavior to the local content characteristics while keeping the processing overhead manageable through simple conditional logic.
Data Source
AI summary
Provided is a method and apparatus for performing intra-prediction using an adaptive filter. The method for performing intra-prediction includes the steps of: determining whether or not to apply a first filter for a reference pixel value on the basis of information of a neighboring block of a current block; applying the first filter for the reference pixel value when it is determined to apply the first filter; performing intra-prediction on the current block on the basis of the reference pixel value; determining whether or not to apply a second filter for a prediction value according to each prediction mode of the current block, which is predicted by the intra-prediction performance on the basis of the information of the neighboring block; and applying the second filter for the prediction value according to each prediction mode of the current block when it is determined to apply the second filter.


