Video Intra Prediction Using Multiple Reference Sample Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding/decoding technologies face inefficiencies in performing intra prediction, particularly in handling high-resolution and stereographic image content, leading to increased data transmission and storage costs.
Innovation Solution
A method and apparatus for video signal processing that utilizes multiple reference sample lines, including adjacent left/right and upper/lower samples, to determine an intra prediction mode and calculate prediction values based on weighted sums or averages, optimizing the encoding/decoding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional intra prediction using only adjacent reference samples is used, then the encoding/decoding process is simple, but the prediction accuracy for high-resolution and stereographic content is insufficient
Solution Approach 1:
The patent divides the reference samples into multiple separate reference sample lines (first reference sample line, second reference sample line, etc.) instead of using a single adjacent reference line. This segmentation allows the encoder to select from multiple non-adjacent reference samples, improving prediction accuracy for high-resolution and stereographic content while managing complexity through structured organization of reference samples.
Solution Approach 2:
The patent extends the reference sample selection from a single adjacent line (one-dimensional approach) to multiple non-adjacent lines (multi-dimensional approach). By utilizing reference samples from different spatial distances and positions (top, left, and diagonally adjacent blocks), the system adds dimensional diversity to the prediction process, enhancing accuracy without proportionally increasing complexity.
2Productivity
If multiple non-adjacent reference samples are used for intra prediction, then the prediction efficiency is improved, but the computational complexity increases
Solution Approach 1:
The patent implements dynamic selection of reference sample lines based on the intra prediction mode. Different reference sample lines are selected depending on the prediction direction (e.g., vertical, horizontal, diagonal modes). This dynamic approach allows the system to use multiple non-adjacent reference samples when beneficial while falling back to simpler single-line selection when appropriate, balancing efficiency improvements with computational complexity.
Solution Approach 2:
The patent changes the parameter of reference sample selection from fixed (only adjacent samples) to variable (multiple non-adjacent samples selected based on mode and position). By introducing mode-dependent selection parameters and distance-based weighting, the system optimizes encoding efficiency while controlling complexity through parameterized decision rules rather than exhaustive processing.
3Manufacturing precision
If only adjacent reference samples are used, then the encoding/decoding process is computationally simple, but the quality of prediction for high-resolution images deteriorates
Solution Approach 1:
The patent performs preliminary organization of reference samples into multiple predefined reference sample lines during the encoding/decoding process. By pre-structuring the reference samples from top, left, and diagonal neighbors into organized lines, the system prepares multiple candidate references in advance, enabling faster selection and improving prediction quality without proportionally increasing processing time during the actual prediction step.
Data Source
AI summary
According to the present invention, there is provided a method of decoding an image, the method including: determining an intra prediction mode of a current block; determining a reference sample line for a prediction target sample included in the current block; and obtaining a prediction value of the prediction target sample on the basis of the intra prediction mode and the reference sample line.