Context Matters: Leveraging Spatiotemporal Metadata for Semi-Supervised Learning on Remote Sensing Images

arXiv:2404.18583v1 Announce Type: new
Abstract: Remote sensing projects typically generate large amounts of imagery that can be used to train powerful deep neural networks. However, the amount of labeled images is often small, as remote sensing applications generally require expert labelers. Thus, semi-supervised learning (SSL), i.e., learning with a small pool of labeled and a larger pool of unlabeled data, is particularly useful in this domain. Current SSL approaches generate pseudo-labels from model predictions for unlabeled samples. As the quality …

abstract applications arxiv context cs.cv expert generate however images metadata networks neural networks projects semi-supervised semi-supervised learning sensing small ssl supervised learning train type