Learning Flexible Generalization in Video Quality Assessment by Bringing Device and Viewing Condition Distributions

ICLR 2026 Conference Submission25543 Authors

Keywords: video quality assessment, subjective dataset, robust perceptual models, human-centered machine learning

Abstract: Video quality assessment (VQA) plays a critical role in optimizing video delivery systems. While numerous objective metrics have been proposed to approximate human perception, the perceived quality strongly depends on viewing conditions and display characteristics. Factors such as ambient lighting, display brightness, and resolution significantly influence the visibility of distortions. In this work, we address the question of the multu-screen quality assessment on mobile devices, as this area still tends to be undercovered. We introduce a large-scale subjective dataset collected across more than 200 Android devices, accompanied by metadata on viewing conditions and display properties. We propose a strategy for aggregated score extraction and adaptation of VQA models to device-specific quality estimation. Our results demonstrate that incorporating device and context information enables more accurate and flexible quality prediction, offering new opportunities for fine-grained optimization in streaming services. We view device and condition variability as a form of natural distributions, and our approach provides a pathway to more robust perceptual quality prediction. Ultimately, this work advances the development of perceptual quality models that bridge the gap between laboratory evaluations and the diverse conditions of real-world media consumption.

Primary Area: datasets and benchmarks

Submission Number: 25543