Adaptive Intra-Prediction for Converging Lines in Video Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current intra-prediction modes in video codecs are inadequate for encoding blocks with converging lines, such as parallel lines viewed from a non-90 degree angle, leading to suboptimal prediction blocks and increased residual errors.
Innovation Solution
The implementation of novel intra-prediction modes that use peripheral pixels along different directions for each prediction pixel, allowing for non-parallel prediction lines and incorporating focal points to determine pixel values through interpolation and warping functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional intra-prediction modes use parallel prediction lines, then the encoding process is simple, but the prediction accuracy is insufficient for scenes with converging lines
Solution Approach 1:
The patent applies dynamics by making the prediction lines adaptive rather than static. The prediction model dynamically adjusts the orientation of prediction lines based on detected edge directions in the current block, allowing the system to transition from simple parallel lines to converging lines that match the actual content structure, thereby improving prediction accuracy without requiring a completely new complex model
Solution Approach 2:
The patent changes the parameter of prediction line orientation from fixed parallel to variable converging lines. By introducing orientation parameters that can be adjusted based on edge detection results, the system modifies the geometric parameters of the prediction model to better fit scenes with converging lines, resolving the contradiction between simplicity and accuracy
2Adaptability or versatility
If conventional intra-prediction uses fixed prediction modes, then the encoding is efficient, but it cannot adapt to different scene structures like converging lines
Solution Approach 1:
The patent segments the prediction process into distinct stages: edge detection, orientation determination, and prediction generation. By dividing the task into separate modules, the system can adapt to different scene structures through the edge detection and orientation determination stages while maintaining efficient prediction generation, thus improving adaptability without significantly compromising coding efficiency
Solution Approach 2:
The patent performs preliminary edge detection and orientation determination before generating the actual prediction blocks. This preliminary action prepares the system to adapt to different scene structures by identifying the appropriate prediction line orientations in advance, allowing the main prediction process to proceed efficiently with pre-determined parameters
Data Source
AI summary
A method for intra-prediction of a current block includes selecting peripheral pixels of the current block, where the peripheral pixels are used to generate a prediction block for the current block; for each prediction pixel of the prediction block, performing steps including selecting two respective pixels of the peripheral pixels; and calculating the prediction pixel by interpolating at least the two respective pixels; and coding a residual block corresponding to a difference between the current block and the prediction block.