Weighted Error Matrix for Video Quality Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video quality measurement tools fail to accurately reflect end-user perception by not considering the perceptual impact of image artifacts, as they treat artifacts of similar intensity equally regardless of their location on the screen, which is critical for providing meaningful video quality information.
Innovation Solution
A video quality measurement tool that performs a video quality test, generates a weighted error matrix based on factors like motion, luminance contrast, and object importance, and updates the score to reflect the impact of artifacts on user perception, providing a weighted video quality score that differentiates between artifacts based on their location and importance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video quality measurement tools treat all artifacts of similar intensity equally, then the measurement process is simple and fast, but the measurement accuracy does not reflect end user perception
Solution Approach 1:
The patent applies local quality by differentiating the weightings of artifacts based on their spatial location within the video frame. Artifacts in perceptually important regions (such as center of screen, high-contrast areas) receive higher weightings than artifacts in less important regions. This allows the measurement tool to reflect end user perception more accurately by treating artifacts differently according to their local context rather than applying uniform treatment to all artifacts of the same intensity.
2Measurement precision
If the video quality measurement tool considers perceptual impact of artifact location, then the measurement reflects end user experience better, but the computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing weighting factors for different spatial regions before actual video quality measurement. The measurement tool uses these pre-computed weightings to quickly assess artifacts without performing complex real-time analysis. This allows the system to reflect perceptual impact of artifact location while minimizing computational time during actual measurement operations.
Data Source
AI summary
A method for performing a video quality measurement which includes performing a video quality test on a frame of a user observed video clip to provide a video quality score, generating a weighted error matrix for the frame, and updating the video quality score based upon the weighted error matrix to provide a weighted video quality score is disclosed.