Work

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

Year: 2024

Type: conference-paper

Abstract: 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.

Source: Proceedings - International Conference on Image Processing

DOI: 10.1109/icip51287.2024.10647697

Authors: Priscilla Indira Osa, Josiane Zerubia, Zoltan Kato

Institutions: University of Genoa, Institut national de recherche en sciences et technologies du numérique, University of Szeged

Language: English

Topic: Remote-Sensing Image Classification

Subfield: Media Technology

Field: Engineering

Domain: Physical Sciences

SDG: Sustainable cities and communities

Open Access status: green

Funder: Szegedi Tudományegyetem