| Item Type |
Conference or Workshop Item
(Paper)
|
| Abstract |
Self-supervised learning has great potential for the remote sensing domain, where unlabelled observations are abundant, but labels are hard to obtain. This work leverages unlabelled multi-modal remote sensing data for augmentation-free contrastive self-supervised learning. Deep neural network models are trained to maximize the similarity of latent representations obtained with different sensing techniques from the same location, while distinguishing them from other locations. We showcase this idea with two self-supervised data fusion methods and compare against standard supervised and self-supervised learning approaches on a land-cover classification task. Our results show that contrastive data fusion is a powerful self-supervised technique to train image encoders that are capable of producing meaningful representations: Simple linear probing performs on par with fully supervised approaches and fine-tuning with as little as 10% of the labelled data results in higher accuracy than supervised training on the entire dataset. |
| Authors |
Scheibenreif, Linus Mathias; Mommert, Michael & Borth, Damian |
| Research Team |
AIML Lab |
| Journal or Publication Title |
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Language |
English |
| Subjects |
computer science |
| HSG Classification |
contribution to scientific community |
| Date |
2022 |
| Publisher |
ISPRS |
| Volume |
V-3-2022 |
| Page Range |
705-711 |
| Number of Pages |
6 |
| Event Title |
ISPRS Congress |
| Event Location |
Nice |
| Event Dates |
6-11 June 2022 |
| Depositing User |
Prof. Dr. Michael Mommert
|
| Date Deposited |
22 Jun 2022 10:26 |
| Last Modified |
17 Aug 2022 17:05 |
| URI: |
https://www.alexandria.unisg.ch/publications/266528 |
Contrastive Self-Supervised Data Fusion for Satellite Imagery
![[img]](http://www.alexandria.unisg.ch/style/images/fileicons/oa_orange.png)
