Absolute Perceptual Video Quality Prediction Across Viewing Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video quality assessment techniques fail to accurately predict perceived quality across diverse viewing devices, as they typically estimate quality relative to the source rather than reflecting human perception on different devices and resolutions.
Innovation Solution
A computer-implemented method that determines absolute quality scores for encoded video content by training source models based on spatial resolutions and generating device equations for various viewing devices, allowing for accurate prediction of perceived quality on target devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If perceptive quality metrics are computed based on subjective viewer ratings, then automated quality assessment can be implemented, but the metrics only estimate quality relative to the source and do not reflect human perception on different devices
Solution Approach 1:
The patent transforms the quality assessment from relative metrics to absolute metrics by introducing device-specific parameters. The system computes absolute quality scores that are adjusted based on viewing device characteristics such as screen resolution, size, and type, thereby changing the parameter space to include device-specific factors that reflect actual human perception on different devices.
Solution Approach 2:
The patent segments the quality assessment process into device-specific evaluations. Instead of a single universal metric, the system creates separate quality score computations for different device types (televisions, mobile devices, tablets, wearables), allowing each device category to have tailored quality predictions that match actual viewing conditions.
2Adaptability or versatility
If quality scores are computed for different sources with varying resolutions, then comprehensive quality evaluation is possible, but scores do not reflect actual perceived quality differences
Solution Approach 1:
The patent introduces absolute quality scores that incorporate source resolution parameters into the assessment. By adjusting the quality metrics to account for source characteristics and mapping them to device-specific perceptions, the system transforms relative quality comparisons into absolute quality predictions that accurately reflect human perception across different source and device combinations.
3Manufacturing precision
If encoding quality is optimized for one device type, then quality can be maximized for that device, but quality consistency across different viewing devices deteriorates
Solution Approach 1:
The patent segments the encoding optimization process into device-specific quality targets. By computing separate absolute quality scores for different device types, the system enables independent optimization for each device category while maintaining awareness of cross-device performance, allowing encoders to adjust parameters to achieve desired quality levels on specific devices without completely sacrificing other device experiences.
Data Source
AI summary
In various embodiments, a perceptual quality application determines an absolute quality score for encoded video content viewed on a target viewing device. In operation, the perceptual quality application determines a baseline absolute quality score for the encoded video content viewed on a baseline viewing device. Subsequently, the perceptual quality application determines that a target value for a type of the target viewing device does not match a base value for the type of the baseline viewing device. The perceptual quality application computes an absolute quality score for the encoded video content viewed on the target viewing device based on the baseline absolute quality score and the target value. Because the absolute quality score is independent of the viewing device, the absolute quality score accurately reflects the perceived quality of a wide range of encoded video content when decoded and viewed on a viewing device.