NovisVQ: A Streaming Convolutional Neural Network for No-Reference Opinion-Unaware Frame Quality Assessment

TL;DR

NovisVQ introduces a scalable, streaming convolutional neural network that assesses video quality without references or opinion labels, effectively leveraging temporal information to outperform existing image-based and opinion-aware methods.

Contribution

The paper presents a novel streaming, no-reference, opinion-unaware VQA model that uses synthetic degradations and temporal modeling to predict full-reference metrics directly from degraded videos.

Findings

Outperforms image-based baseline in generalization across degradations

Achieves higher correlation with full-reference metrics than BRISQUE

Validates the importance of temporal modeling for video quality assessment

Abstract

Video quality assessment (VQA) is vital for computer vision tasks, but existing approaches face major limitations: full-reference (FR) metrics require clean reference videos, and most no-reference (NR) models depend on training on costly human opinion labels. Moreover, most opinion-unaware NR methods are image-based, ignoring temporal context critical for video object detection. In this work, we present a scalable, streaming-based VQA model that is both no-reference and opinion-unaware. Our model leverages synthetic degradations of the DAVIS dataset, training a temporal-aware convolutional architecture to predict FR metrics (LPIPS , PSNR, SSIM) directly from degraded video, without references at inference. We show that our streaming approach outperforms our own image-based baseline by generalizing across diverse degradations, underscoring the value of temporal modeling for scalable VQA in real-world vision systems. Additionally, we demonstrate that our model achieves higher correlation with full-reference metrics compared to BRISQUE, a widely-used opinion-aware image quality assessment baseline, validating the effectiveness of our temporal, opinion-unaware approach.