Jamali, Mohammadreza, Coulombe, Stéphane, Vakili, Ahmad et Vazquez, Carlos.
2020.
« LSTM-based viewpoint prediction for multi-quality tiled video coding in virtual reality streaming ».
In IEEE International Symposium on Circuits and Systems (ISCAS) (Seville, Spain, Oct. 10-21, 2020)
Institute of Electrical and Electronics Engineers.
Compte des citations dans Scopus : 15.
Prévisualisation |
PDF
Coulombe-S-2020-22700-1.pdf - Version acceptée Licence d'utilisation : Tous les droits réservés aux détenteurs du droit d'auteur. Télécharger (418kB) | Prévisualisation |
Résumé
Virtual reality (VR) streaming is impaired by the large amount of data required to deliver 360-degree video resulting in low-quality end user experience when network bandwidth is limited, or latency is high. To address these challenges, proposed in this paper is a novel method for viewpoint prediction for long-term horizons in VR streaming. This method uses a long short-term memory (LSTM) encoder-decoder network to carry out a sequence-to-sequence prediction. To enhance the results obtained by this network, experiments are performed using viewpoint information from users on low-latency networks. By applying an effective tile-based quality assignment after viewpoint prediction, a 61% average bandwidth reduction, with respect to the transmission of the whole ERP frame, is achieved along with a high-quality viewport rendered to the end user.
| Type de document: | Compte rendu de conférence |
|---|---|
| ISBN: | 2158-1525 |
| Professeur: | Professeur Coulombe, Stéphane Vázquez, Carlos |
| Affiliation: | Génie logiciel et des technologies de l'information, Génie logiciel et des technologies de l'information |
| Date de dépôt: | 03 juin 2021 13:35 |
| Dernière modification: | 21 déc. 2022 15:32 |
| URI: | https://espace2.etsmtl.ca/id/eprint/22700 |
Actions (Authentification requise)
 |
Dernière vérification avant le dépôt |