Multi-Task Learning of Height and Semantics from Aerial Images

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 .

Published IEEE Geoscience and Remote Sensing Letters. Code https://github.com/marcelampc/mtl_aerial_images

References in corpus (1)

• Fully Convolutional Networks for Semantic Segmentation

Cited by in corpus (3)

• Geometry-Aware Segmentation of Remote Sensing Images via Implicit Height Estimation
• Focusing on Shadows for Predicting Heightmaps from Single Remotely Sensed RGB Images with Deep Learning
• Mapping Vulnerable Populations with AI