Video Encoder Pre-Analysis Using Pseudo-Affine Motion Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The difference in inter-frame prediction results between pre-analysis modules and primary encoder modules in video coding, particularly due to the use of conventional single-motion-vector methods in pre-analysis modules, leads to inefficiencies and inaccurate guidance for newer video coding standards like VVC.
Innovation Solution
Implement a pseudo-affine transformation mode in pre-analysis modules to mimic the prediction manner of the primary encoder's affine transformation mode, allowing for more accurate inter-frame prediction by comparing costs between single-motion-vector and pseudo-affine transformation modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional single-motion-vector mode is used in pre-analysis modules, then device complexity is reduced, but measurement precision of inter-frame prediction deteriorates
Solution Approach 1:
The patent segments the prediction process into two stages: pre-analysis stage using simplified single-motion-vector mode for speed, and final encoding stage using comprehensive affine transformation mode for accuracy. This segmentation allows each stage to optimize for its specific goal without compromising the other.
Solution Approach 2:
The pre-analysis module performs preliminary motion estimation using single-motion-vector mode to generate initial prediction results and costs. These preliminary results serve as a foundation for the subsequent affine transformation optimization, enabling the system to quickly eliminate obviously suboptimal modes before detailed analysis.
2Measurement precision
If pseudo-affine transformation mode is implemented in pre_analysis, then prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies pseudo-affine transformation locally only when the pre-analysis module detects potential affine motion characteristics, rather than universally applying it to all blocks. This localized application maintains accuracy for complex motions while avoiding unnecessary complexity for simple translational motions.
Solution Approach 2:
The system dynamically selects between single-motion-vector mode and pseudo-affine transformation mode based on the characteristics of each video block. The pre-analysis results guide this dynamic selection, allowing the system to adapt its complexity level to the actual motion content of each region.
3Productivity
If single-motion-vector mode is used for inter-frame prediction, then processing speed is maintained, but prediction cost accuracy deteriorates
Solution Approach 1:
The patent introduces an intermediary pseudo-affine transformation step that bridges the gap between simple single-motion-vector prediction and complex affine transformation. This intermediary approach provides more accurate cost calculations than single-motion-vector mode while avoiding the full computational burden of complete affine transformation optimization.
Data Source
AI summary
The application discloses an image pre-analysis method, applied to a pre-analysis module of an encoder. The method includes: performing downsampling on a to-be-processed image, and dividing the to-be-processed image into square blocks of a same size; performing inter-frame prediction on a current block based on a single-motion-vector mode, to determine a best cost and a corresponding prediction direction; performing inter-frame prediction on the current block based on a pseudo-affine transformation mode by using the prediction direction as a search direction of the pseudo-affine transformation mode, to obtain an affine transformation cost; and comparing a value of the best cost with a value of the affine transformation cost, and determining, based on a comparison result, a best mode for performing pre-analysis inter-frame prediction on the current block. The application further discloses an image pre-analysis system, an electronic apparatus, and a computer-readable storage medium.