Bi-Directional Video Prediction Using Gradient-Based Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video codecs struggle with efficient encoding and decoding of high-resolution or high-quality video content due to limitations in encoding methods based on tree structures, leading to inefficient data compression and prediction.
Innovation Solution
Implement bi-directional motion prediction mode that utilizes gradient values of reference blocks to generate prediction pixel values by applying interpolation filters to neighboring regions, enhancing encoding and decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional tree-structure encoding methods are used, then device complexity is reduced, but encoding and decoding efficiency deteriorates
Solution Approach 1:
The patent changes the parameter of prediction by introducing gradient values and applying interpolation filters to generate sub-pixel precision prediction values. This transforms the conventional integer-pixel prediction into sub-pixel prediction, improving prediction accuracy and thus encoding efficiency without fundamentally changing the tree-structure encoding framework
Solution Approach 2:
The patent performs preliminary gradient calculation and interpolation filtering on reference blocks before generating prediction values. By pre-calculating gradient values and applying interpolation filters in advance, the method prepares high-precision prediction data that improves subsequent encoding efficiency without adding complex real-time processing requirements
2Measurement precision
If gradient values and interpolation filters are applied to generate prediction pixel values, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies gradient calculation and interpolation filtering locally to reference blocks and their neighboring regions rather than globally to entire pictures. This localized approach improves prediction accuracy at specific block boundaries while limiting the computational increase to only the necessary local areas
Solution Approach 2:
The patent applies interpolation filters to generate sub-pixel precision values (excessive action) only where needed for prediction, rather than processing all pixels uniformly. This partial application of complex operations achieves higher prediction accuracy without proportionally increasing overall computational complexity
3Measurement precision
If bi-directional motion prediction mode is used, then prediction quality is improved, but data volume increases
Solution Approach 1:
The patent changes the prediction quality parameter by using gradient-based sub-pixel interpolation in bi-directional prediction mode. This transforms standard pixel-level prediction into sub-pixel precision prediction, improving prediction quality and reducing residual data volume even though more reference data is initially processed
Data Source
AI summary
A video decoding and video encoding method of performing inter prediction in a bi-directional motion prediction mode, in which a prediction pixel value of a current block may be generated by not only using a pixel value of a first reference block of a first reference picture and a pixel value of a second reference block of a second reference picture, but also using a first gradient value of the first reference block and a second gradient value of the second reference block, in a bi-directional motion prediction mode. Accordingly, encoding and decoding efficiency may be increased since a prediction block similar to an original block may be generated.