Communication Dans Un Congrès Année : 2025

UNETRSal: Saliency Prediction with Hybrid Transformer-Based Architecture

Résumé

Saliency prediction plays a critical role in understanding visual attention as it is a cornerstone for both natural scene understanding and automated document analysis. In this work, we propose the UN-ETRSal model for saliency prediction. Based on UNETR transformerbased model, we introduce a new decoder to increase efficiency on 2D images. Comprehensive evaluations on benchmark datasets, such as SAL-ICON and CAT2000, demonstrate that UNETRSal achieves state-ofthe-art performance across multiple saliency metrics, surpassing both conventional CNN-based and transformer-based methods. These results not only underscore the strengths of hybrid transformer architectures in modeling visual attention but also highlight the potential impact on advancing document representation modeling and layout analysis.

Fichier principal
Vignette du fichier
ACIVS_2025.pdf (810.76 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05110284 , version 1 (12-06-2025)

Licence

Identifiants

  • HAL Id : hal-05110284 , version 1

Citer

Azamat Kaibaldiyev, Jeremie Pantin, Alexis Lechervy, Fabrice Maurel, Youssef Chahir, et al.. UNETRSal: Saliency Prediction with Hybrid Transformer-Based Architecture. Advanced Concepts for Intelligent Vision Systems (ACIVS), Jul 2025, Tokyo (JP), Japan. ⟨hal-05110284⟩
617 Consultations
315 Téléchargements

Partager

  • More