Multi-Task Learning of Height and Semantics from Aerial Images

November 18, 2019 · Entered Twilight · 🏛 IEEE Geoscience and Remote Sensing Letters

"Last commit was 6.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: LICENSE.md, README.md, bayesian_test_raster.py, config, dataloader, images, main_raster.py, models, networks, options, std_scripts, util

Authors Marcela Carvalho, Bertrand Le Saux, Pauline Trouvé-Peloux, Frédéric Champagnat, Andrés Almansa arXiv ID 1911.07543 Category cs.CV: Computer Vision Cross-listed cs.LG Citations 59 Venue IEEE Geoscience and Remote Sensing Letters Repository https://github.com/marcelampc/mtl_aerial_images ⭐ 85 Last Checked 1 month ago

Abstract

Aerial or satellite imagery is a great source for land surface analysis, which might yield land use maps or elevation models. In this investigation, we present a neural network framework for learning semantics and local height together. We show how this joint multi-task learning benefits to each task on the large dataset of the 2018 Data Fusion Contest. Moreover, our framework also yields an uncertainty map which allows assessing the prediction of the model. Code is available at https://github.com/marcelampc/mtl_aerial_images .