Image Prediction Using Reference Pixels and Affine Candidates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies face challenges in efficiently performing intra prediction and inter prediction, particularly in determining affine candidates and motion compensation, which affect encoding/decoding performance and accuracy.
Innovation Solution
The method involves checking reference pixel regions for correlation information, determining reference pixel processing settings, generating candidate lists for motion information prediction, deriving control point vectors, and performing intra and inter prediction based on these settings, including spatial, temporal, and constructed candidates, with motion vectors derived in units of sub-blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion compensation is performed in units of sub-blocks, then inter prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The current block is divided into multiple sub-blocks, and motion compensation is performed independently for each sub-block. This segmentation allows for more precise motion estimation at the sub-block level, improving inter prediction accuracy while distributing the computational complexity across multiple smaller units rather than processing the entire block as one unit.
2Reliability
If multiple affine candidates are generated and evaluated, then coding performance is improved, but encoding time increases
Solution Approach 1:
Multiple affine motion candidates are pre-generated and stored in a candidate list before the actual encoding process. These candidates include spatial candidates from neighboring blocks, temporal candidates from co-located blocks in reference pictures, and constructed candidates. By preparing these candidates in advance, the encoder can quickly select from pre-computed options during encoding, improving coding performance while reducing the real-time computational burden.
Solution Approach 2:
Instead of exhaustively searching all possible motion vectors, the method generates a limited set of affine candidates that are most likely to be optimal. This partial action approach generates sufficient candidates to achieve high coding performance without the excessive computational cost of exhaustive search, balancing encoding time and coding performance.
3Measurement precision
If reference pixel processing is performed with strict availability checking, then intra prediction accuracy is improved, but processing complexity increases
Solution Approach 1:
The reference pixel processing method dynamically adapts to the availability of reference pixels at block boundaries. The system checks whether reference pixels are available and adjusts the prediction process accordingly, selecting appropriate reference pixel lines and processing methods based on the specific boundary conditions. This dynamic adaptation maintains high intra prediction accuracy while avoiding unnecessary processing complexity in cases where reference pixels are unavailable.
Data Source
AI summary
An image prediction method according to the present invention may comprise: identifying a reference pixel region designated for obtaining correlation information; determining a reference pixel processing configuration on the basis of determination of the availability of the reference pixel region; and performing intra prediction according to the determined reference pixel processing. As described above, performing intra prediction on the basis of the availability of a reference pixel according to the present invention can improve encoding performance.