MEMC Parameter Calibration for Consistent Video Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing MEMC algorithms in different video processing devices often produce inconsistent compensation results due to variations in algorithms across generations or product lines, leading to differing digital video outputs.
Innovation Solution
A motion estimation and motion compensation system with a correction function that includes a storage unit, MEMC unit, and processing unit to calibrate parameters by comparing default and testing feature sets, using linear regression and least squares methods to generate calibration parameters for consistent output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MEMC algorithms are iteratively updated across different generations or product lines, then the video processing capabilities are improved, but the compensation results become inconsistent across different devices
Solution Approach 1:
The patent applies parameter changes by introducing calibration parameters that adjust the MEMC algorithm's behavior based on device-specific characteristics. The system calculates difference values between default feature sets and testing feature sets, then generates calibration parameters through linear regression to modify the algorithm's output, ensuring consistent compensation results across different device generations while maintaining improved video processing capabilities.
Solution Approach 2:
The patent implements feedback by comparing the actual compensation results (testing feature set) against expected results (default feature set) and using the difference values to generate calibration parameters. This closed-loop feedback mechanism allows the system to automatically adjust and correct inconsistencies in MEMC compensation across different devices, ensuring reliability while allowing algorithm iterations for productivity improvement.
2Reliability
If manual calibration of MEMC parameters is performed for each device, then compensation result consistency is improved, but developer workload and development costs increase
Solution Approach 1:
The patent applies self-service by enabling the MEMC system to automatically calibrate itself through the calibration process. The system autonomously fetches testing feature sets, calculates difference values, performs linear regression to generate calibration parameters, and applies corrections without requiring manual intervention from developers. This automation maintains compensation result consistency while significantly reducing developer workload and development complexity.
Solution Approach 2:
The patent replaces the mechanical/manual calibration process with an automated computational system. Instead of requiring developers to manually adjust parameters for each device, the system uses algorithms (linear regression, least squares) to automatically calculate and apply calibration parameters, substituting manual mechanical adjustment with automated digital processing.
Data Source
AI summary
A motion estimation and motion compensation (MEMC) system with calibration function and a parameter calibration method thereof relate to a parameter calibration method for MEMC. The parameter calibration method is configured to calibrate an MEMC program. The parameter calibration method includes storing a default feature set; performing the MEMC program on a calibrated video to fetch a testing feature set; generating a calibration parameter according to a difference vale between the default feature set and the testing feature set; and adjusting a correction parameter of the MEMC calibration program in accordance with the calibration parameter.