OTT Video Quality Analysis Using Precomputed VMAF Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is no appropriate methodology to perform reference-based Video Quality Analysis (VQA) for live or VOD internet streams due to lack of access to source streams, varying bitrates, and absence of automated workflows.
Innovation Solution
Pre-calculating Video Multimethod Assessment Fusion (VMAF) scores for media and correlating them with actual playback data to enhance video quality analysis (VQAaaS) for OTT streaming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reference-based VQA is performed for OTT streaming, then video quality measurement accuracy is improved, but access to source streams is lost due to compression and re-encoding
Solution Approach 1:
The system performs preliminary VMAF quality analysis on the source video stream before it undergoes compression and re-encoding for OTT distribution. By calculating reference quality metrics in advance and storing them alongside the encoded segments, the system preserves ground truth quality data even though the original source stream is no longer accessible during playback. This preliminary action enables accurate quality assessment of the compressed stream by comparing against pre-calculated reference metrics.
2Adaptability or versatility
If automated VQA workflow is implemented, then video quality analysis capability is improved, but system complexity increases
Solution Approach 1:
The system implements a universal automated workflow that handles multiple video quality assessment scenarios (pre-encoding, post-encoding, and runtime quality monitoring) through a single integrated VMAF-based framework. This multi-functional approach consolidates what would otherwise require separate analysis systems for different stages of the streaming pipeline, reducing overall system complexity while maintaining broad video quality analysis capability across various OTT scenarios.
3Measurement precision
If VMAF scores are calculated for each segment, then video quality analysis precision is improved, but computational time increases
Solution Approach 1:
The system pre-calculates VMAF quality scores for video segments during the encoding and packaging phase, before the segments are distributed through the CDN. These pre-computed quality metrics are stored as metadata alongside the video segments. During playback, the quality analysis is instantaneous as it simply retrieves and displays the pre-calculated scores, eliminating the need for real-time quality computation and thus avoiding computational time delays.
Data Source
AI summary
This disclosure provides for automated techniques to measure full reference-based QoE or VQA-as-a-Service (VQAaaS) for an Internet video stream. Generally, the approach herein involves pre-calculating VMAF scores for given media and then correlating those scores with VMAF scores computed from actual playback segments for the given media. By leveraging the pre-calculated VMAF scores and correlating them with playback data, the system provides for enhanced and accurate video quality analysis (VQA) to enable optimization of viewer Quality of Experience (QoE).