Sphere-GAN: a GAN-based Approach for Saliency Estimation in 360° Videos

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wahba, Mahmoud Z. A., Baldoni, Sara, Battisti, Federica
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915495337263104
author Wahba, Mahmoud Z. A.
Baldoni, Sara
Battisti, Federica
author_facet Wahba, Mahmoud Z. A.
Baldoni, Sara
Battisti, Federica
contents The recent success of immersive applications is pushing the research community to define new approaches to process 360° images and videos and optimize their transmission. Among these, saliency estimation provides a powerful tool that can be used to identify visually relevant areas and, consequently, adapt processing algorithms. Although saliency estimation has been widely investigated for 2D content, very few algorithms have been proposed for 360° saliency estimation. Towards this goal, we introduce Sphere-GAN, a saliency detection model for 360° videos that leverages a Generative Adversarial Network with spherical convolutions. Extensive experiments were conducted using a public 360° video saliency dataset, and the results demonstrate that Sphere-GAN outperforms state-of-the-art models in accurately predicting saliency maps.
format Preprint
id arxiv_https___arxiv_org_abs_2509_11948
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sphere-GAN: a GAN-based Approach for Saliency Estimation in 360° Videos
Wahba, Mahmoud Z. A.
Baldoni, Sara
Battisti, Federica
Computer Vision and Pattern Recognition
Multimedia
Image and Video Processing
The recent success of immersive applications is pushing the research community to define new approaches to process 360° images and videos and optimize their transmission. Among these, saliency estimation provides a powerful tool that can be used to identify visually relevant areas and, consequently, adapt processing algorithms. Although saliency estimation has been widely investigated for 2D content, very few algorithms have been proposed for 360° saliency estimation. Towards this goal, we introduce Sphere-GAN, a saliency detection model for 360° videos that leverages a Generative Adversarial Network with spherical convolutions. Extensive experiments were conducted using a public 360° video saliency dataset, and the results demonstrate that Sphere-GAN outperforms state-of-the-art models in accurately predicting saliency maps.
title Sphere-GAN: a GAN-based Approach for Saliency Estimation in 360° Videos
topic Computer Vision and Pattern Recognition
Multimedia
Image and Video Processing
url https://arxiv.org/abs/2509.11948