No-reference video quality assessment using distortion learning and temporal attention

Kossi, Koffi, Coulombe, Stéphane, Desrosiers, Christian et Gagnon, Ghyslain. 2022. « No-reference video quality assessment using distortion learning and temporal attention ». IEEE Access, vol. 10. pp. 41010-41022.
Compte des citations dans Scopus : 4.

Prévisualisation

PDF Coulombe-S-2022-24440.pdf - Version publiée Licence d'utilisation : Creative Commons CC BY. Télécharger (1MB) | Prévisualisation

Résumé

The rapid growth of video consumption and multimedia applications has increased the interest of the academia and industry in building tools that can evaluate perceptual video quality. Since videos might be distorted when they are captured or transmitted, it is imperative to develop reliable methods for no-reference video quality assessment (NR-VQA). To date, most NR-VQA models in prior art have been proposed for assessing a specific category of distortion, such as authentic distortions or traditional distortions. Moreover, those developed for both authentic and traditional distortions video databases have so far led to poor performances. This resulted in the reluctance of service providers to adopt multiple NR-VQA approaches, as they prefer a single algorithm capable of accurately estimating video quality in all situations. Furthermore, many existing NR-VQA methods are computationally complex and therefore impractical for various real-life applications. In this paper, we propose a novel deep learning method for NR-VQA based on multi-task learning where the distortion of individual frames in a video and the overall quality of the video are predicted by a single neural network. This enables to train the network with a greater amount and variety of data, thereby improving its performance in testing. Additionally, our method leverages temporal attention to select the frames of a video sequence which contribute the most to its perceived quality. The proposed algorithm is evaluated on five publicly-available video quality assessment (VQA) databases containing traditional and authentic distortions. Results show that our method outperforms the state-of-the- art on traditional distortion databases such as LIVE VQA and CSIQ video, while also delivering competitive performance on databases containing authentic distortions such as KoNViD-1k, LIVE-Qualcomm and CVD2014.

Type de document:	Article publié dans une revue, révisé par les pairs
Professeur:	Professeur Coulombe, Stéphane Desrosiers, Christian Gagnon, Ghyslain
Affiliation:	Génie logiciel et des technologies de l'information, Génie logiciel et des technologies de l'information, Génie électrique
Date de dépôt:	02 juin 2022 19:00
Dernière modification:	23 juin 2022 14:58
URI:	https://espace2.etsmtl.ca/id/eprint/24440

Actions (Authentification requise)

Dernière vérification avant le dépôt