In recent years self-supervised learning has emerged as a promising candidate for unsupervised representation learning. In the visual domain its applications are mostly studied in the context of images of natural scenes. However, its applicability is especially interesting in specific areas, like remote sensing and medicine, where it is hard to obt...

Self-Supervised Learning of Remote Sensing Scene Representations Using Contrastive Multiview Coding