PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
September 10, 2025IEEE Transactions on Pattern Analysis and Machine Intelligence

Spherical Vision Transformers for Audio-Visual Saliency Prediction in 360^ Videos

View Full Paper
Ask AI
Bookmark
Share

Authors

MCMert CokelekHOHalit OzsoyNİNevrez İmamoğlu

Discussion

Loading...

Member takes

Overview

Observational analysis examines audio-visual saliency in 360° videos, suggesting spatial audio improves prediction accuracy.

Key Points

  • SalViT360 and SalViT360-AV significantly outperform existing methods in predicting viewer attention in 360° scenes.
  • The new YT360-EyeTracking dataset comprises 81 omnidirectional videos under varying audio-visual conditions.
  • Incorporating spatial audio cues into models is crucial for accurate saliency prediction in ODVs.
  • This research highlights the importance of addressing spherical distortion in video saliency prediction.

Cite This Study

Cokelek et al. (2025) studied this question.

synapsesocial.com/papers/68c1dd9b54b1d3bfb60fc25bhttps://doi.org/10.1109/tpami.2025.3604091
View Full Paper
Ask AI
Bookmark
Share