The recent growth in the number of satellite images fosters the development of effective deep-learning techniques for Remote Sensing (RS). However, their full potential is untapped due to the lack of large annotated datasets. Such a problem is usually countered by fine-tuning a feature extractor that is previously trained on the ImageNet dataset. Unfortunately, the domain of natural images differs from the RS one, which hinders the final performance. In this work, we propose to learn meaningful representations from satellite imagery, leveraging its high-dimensionality spectral bands to reconstruct the visible colors. We conduct experiments on land cover classification (BigEarthNet) and West Nile Virus detection, showing that colorization is a solid pretext task for training a feature extractor. Furthermore, we qualitatively observe that guesses based on natural images and colorization rely on different parts of the input. This paves the way to an ensemble model that eventually outperforms both the above-mentioned techniques.

The color out of space: learning self-supervised representations for Earth Observation imagery / Vincenzi, Stefano; Porrello, Angelo; Buzzega, Pietro; Cipriano, Marco; Fronte, Pietro; Cuccu, Roberto; Ippoliti, Carla; Conte, Annamaria; Calderara, Simone. - (2020), pp. 3034-3041. ((Intervento presentato al convegno 25th International Conference on Pattern Recognition tenutosi a Milan, Italy nel 10-15 January 2021 [10.1109/ICPR48806.2021.9413112].

The color out of space: learning self-supervised representations for Earth Observation imagery

Stefano Vincenzi;Angelo Porrello;Pietro Buzzega;Marco Cipriano;Simone Calderara
2020-01-01

Abstract

The recent growth in the number of satellite images fosters the development of effective deep-learning techniques for Remote Sensing (RS). However, their full potential is untapped due to the lack of large annotated datasets. Such a problem is usually countered by fine-tuning a feature extractor that is previously trained on the ImageNet dataset. Unfortunately, the domain of natural images differs from the RS one, which hinders the final performance. In this work, we propose to learn meaningful representations from satellite imagery, leveraging its high-dimensionality spectral bands to reconstruct the visible colors. We conduct experiments on land cover classification (BigEarthNet) and West Nile Virus detection, showing that colorization is a solid pretext task for training a feature extractor. Furthermore, we qualitatively observe that guesses based on natural images and colorization rely on different parts of the input. This paves the way to an ensemble model that eventually outperforms both the above-mentioned techniques.
25th International Conference on Pattern Recognition
Milan, Italy
10-15 January 2021
3034
3041
Vincenzi, Stefano; Porrello, Angelo; Buzzega, Pietro; Cipriano, Marco; Fronte, Pietro; Cuccu, Roberto; Ippoliti, Carla; Conte, Annamaria; Calderara, Simone
The color out of space: learning self-supervised representations for Earth Observation imagery / Vincenzi, Stefano; Porrello, Angelo; Buzzega, Pietro; Cipriano, Marco; Fronte, Pietro; Cuccu, Roberto; Ippoliti, Carla; Conte, Annamaria; Calderara, Simone. - (2020), pp. 3034-3041. ((Intervento presentato al convegno 25th International Conference on Pattern Recognition tenutosi a Milan, Italy nel 10-15 January 2021 [10.1109/ICPR48806.2021.9413112].
File in questo prodotto:
File Dimensione Formato  
the_color_out_main_paper.pdf

Open access

Descrizione: Articolo principale
Tipologia: Versione dell'autore revisionata e accettata per la pubblicazione
Dimensione 1.62 MB
Formato Adobe PDF
1.62 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1211826
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 13
  • ???jsp.display-item.citation.isi??? 9
social impact