Gabor Feature Network for Transformer-based Building Change Detection Model in Remote Sensing - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

Gabor Feature Network for Transformer-based Building Change Detection Model in Remote Sensing

Résumé

Detecting building change in bitemporal remote sensing (RS) imagery requires a model to highlight the changes in buildings and ignore the irrelevant changes of other objects and sensing conditions. Buildings have comparatively less diverse textures than other objects and appear as repetitive visual patterns on RS images. In this paper, we propose Gabor Feature Network (GFN) to extract the distinctive repetitive texture features of buildings. Furthermore, we also design Feature Fusion Module (FFM) to fuse the extracted multiscale features from GFN with the features from a Transformer-based encoder to pass on the texture features to different parts of the model. Using GFN and FFM, we design a Transformer-based model, called GabFormer for building change detection. Experimental results on the LEVIR-CD and WHU-CD datasets indicate that GabFormer outperforms other SOTA models and in particular show significant improvement in the generalization capability. Our code is available on https://github.com/Ayana-Inria/GabFormer
Fichier principal
Vignette du fichier
ICIP_camera-ready-final.pdf (1.55 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04619245 , version 1 (20-06-2024)

Licence

Identifiants

  • HAL Id : hal-04619245 , version 1

Citer

Priscilla Indira Osa, Josiane Zerubia, Zoltan Kato. Gabor Feature Network for Transformer-based Building Change Detection Model in Remote Sensing. ICIP 2024 - IEEE International Conference on Image Processing, Oct 2024, Abu Dhabi, United Arab Emirates. ⟨hal-04619245⟩

Collections

INRIA INRIA2
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More