Abstract. With the growth of the availability and quality of satellite images, automatic 3D reconstruction from optical satellite images remains a popular research topic. Numerous applications, such as telecommunications and defence, directly benefit from the use of 3D models of both urban and rural scenes. While most of the state-of-the-art methods use stereo pairs for 3D reconstruction, such pairs are not immediately available anywhere in the world. In this paper, we propose an automatic pipeline for very-large-scale 3D reconstruction of urban and rural scenes from one high-resolution satellite image. Convolutional neural networks are trained to extract key semantic information. The extracted information is then converted into GIS vector format, and enriched by both terrain and object height information. The final classification step is applied, yielding a 16-class 3D map. The presented pipeline is operational and available for commercial purposes under the BrightEarth trademark.

Abstract. The understanding of the Earth through global land monitoring from satellite images paves the way towards many applications including flight simulations, urban management and telecommunications. The twin satellites from the Sentinel-2 mission developed by the European Space Agency (ESA) provide 13 spectral bands with a high observation frequency worldwide. In this paper, we present a novel multi-temporal approach for land-cover classification of Sentinel-2 images whereby a time-series of images is classified using fully convolutional network U-Net models and then coupled by a developed probabilistic algorithm. The proposed pipeline further includes an automatic quality control and correction step whereby an external source can be introduced in order to validate and correct the deep learning classification. The final step consists of adjusting the combined predictions to the cloud-free mosaic built from Sentinel-2 L2A images in order for the classification to more closely match the reference mosaic image.