Multi-Sample Intra-Prediction for Video Coding Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding and decoding technologies face inefficiencies and increased complexity, particularly in intra-prediction methods, which affect the encoding/decoding of images with high correlation between samples, leading to reduced coding efficiency.
Innovation Solution
A multi-sample prediction method is introduced, where a sample group is determined within a block, a representative position is identified, and a representative prediction value is calculated based on this position, allowing for efficient prediction of multiple samples using this value, thereby reducing the number of interpolation processes and overall complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If intra-prediction is performed using traditional methods for each sample individually, then prediction accuracy is maintained, but encoding/decoding complexity and processing time increase
Solution Approach 1:
The current block is divided into multiple sample groups, where each group contains multiple samples that share the same prediction value. This segmentation allows the system to process groups rather than individual samples, reducing the number of interpolation operations required while maintaining prediction accuracy for high-correlation regions
Solution Approach 2:
Multiple samples within a sample group are merged by assigning them a single representative prediction value. This merging strategy reduces the total number of prediction calculations needed, directly lowering encoding/decoding complexity while preserving coding efficiency through adaptive group-based prediction
2Measurement precision
If multiple interpolation processes are performed for each sample, then prediction precision is improved, but the number of calculations and processing complexity increase
Solution Approach 1:
Instead of performing full interpolation for every sample, the system applies interpolation selectively at the group level. A single representative prediction value is calculated for each sample group through one interpolation process, then applied to all samples in that group. This partial action approach maintains sufficient prediction precision while dramatically reducing calculation complexity
Data Source
AI summary
The multi sample prediction method of the present invention comprises the steps of: determining a sample group consisting of a plurality of samples inside a decoding target block; determining a representative position corresponding to the sample group, inside the decoding target block; determining a representative prediction value for the sample group, on the basis of the determined representative position; and determining the determined representative prediction value as the final prediction value for each of the plurality of samples making up the sample group. The present invention enhances efficiency in encoding/decoding and reduces complexity thereof.