A contrastive method based on elevation data for remote sensing with scarce and high level semantic labels. (arXiv:2304.06857v1 [cs.CV])

This work proposes a hybrid unsupervised/supervised learning method to
pretrain models applied in earth observation downstream tasks where only a
handful of labels denoting very general semantic concepts are available. We
combine a contrastive approach to pretrain models with a pretext task to
predict spatially coarse elevation maps which are commonly available worldwide.
The intuition behind is that there is generally some correlation between the
elevation and targets in many remote sensing tasks, allowing the model to
pre-learn useful representations. …

arxiv correlation data earth earth observation general hybrid intuition labels learn maps observation remote semantic sensing supervised learning unsupervised work