Land cover maps are a pivotal element in a wide range of Earth Observation (EO) applications. However, anno- tating large datasets to develop supervised systems for re- mote sensing (RS) semantic segmentation is costly and time- consuming. Unsupervised Domain Adaption (UDA) could tackle these issues by adapting a model trained on a source domain, where labels are available, to a target domain, without annotations. UDA, while gaining importance in computer vision, is still under-investigated in RS. Thus, we propose a new lightweight model, GeoMultiTaskNet, based on two contributions: a GeoMultiTask module (GeoMT), which utilizes geographical coordinates to align the source and target domains, and a Dynamic Class Sampling (DCS) strategy, to adapt the semantic segmentation loss to the fre- quency of classes. This approach is the first to use geo- graphical metadata for UDA in semantic segmentation. It reaches state-of-the-art performances (47,22% mIoU), re- ducing at the same time the number of parameters (33M), on a subset of the FLAIR dataset, a recently proposed dataset properly shaped for RS UDA, used for the first time ever for research scopes here.
GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates / Marsocci, Valerio; Gonthier, Nicolas; Garioud, Anatol; Scardapane, Simone; Mallet, Clément. - (2023), pp. 2075-2085. (Intervento presentato al convegno 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) tenutosi a Vancouver, Canada) [10.1109/cvprw59228.2023.00201].
GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates
Marsocci, Valerio
;Scardapane, Simone;
2023
Abstract
Land cover maps are a pivotal element in a wide range of Earth Observation (EO) applications. However, anno- tating large datasets to develop supervised systems for re- mote sensing (RS) semantic segmentation is costly and time- consuming. Unsupervised Domain Adaption (UDA) could tackle these issues by adapting a model trained on a source domain, where labels are available, to a target domain, without annotations. UDA, while gaining importance in computer vision, is still under-investigated in RS. Thus, we propose a new lightweight model, GeoMultiTaskNet, based on two contributions: a GeoMultiTask module (GeoMT), which utilizes geographical coordinates to align the source and target domains, and a Dynamic Class Sampling (DCS) strategy, to adapt the semantic segmentation loss to the fre- quency of classes. This approach is the first to use geo- graphical metadata for UDA in semantic segmentation. It reaches state-of-the-art performances (47,22% mIoU), re- ducing at the same time the number of parameters (33M), on a subset of the FLAIR dataset, a recently proposed dataset properly shaped for RS UDA, used for the first time ever for research scopes here.File | Dimensione | Formato | |
---|---|---|---|
Marsocci_GeoMultiTaskNet_2023.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
7.85 MB
Formato
Adobe PDF
|
7.85 MB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.