3D Building Reconstruction from Monocular Remote Sensing Images with Multi-level Supervisions
This work addresses the problem of scalable 3D building reconstruction for remote sensing applications by reducing annotation costs, though it is incremental in improving existing methods.
The paper tackles 3D building reconstruction from monocular remote sensing images by proposing MLS-BRN, a multi-level supervised network that reduces reliance on expensive 3D annotations, achieving competitive performance with fewer 3D-annotated samples and improving footprint extraction and 3D reconstruction compared to state-of-the-art methods.
3D building reconstruction from monocular remote sensing images is an important and challenging research problem that has received increasing attention in recent years, owing to its low cost of data acquisition and availability for large-scale applications. However, existing methods rely on expensive 3D-annotated samples for fully-supervised training, restricting their application to large-scale cross-city scenarios. In this work, we propose MLS-BRN, a multi-level supervised building reconstruction network that can flexibly utilize training samples with different annotation levels to achieve better reconstruction results in an end-to-end manner. To alleviate the demand on full 3D supervision, we design two new modules, Pseudo Building Bbox Calculator and Roof-Offset guided Footprint Extractor, as well as new tasks and training strategies for different types of samples. Experimental results on several public and new datasets demonstrate that our proposed MLS-BRN achieves competitive performance using much fewer 3D-annotated samples, and significantly improves the footprint extraction and 3D reconstruction performance compared with current state-of-the-art. The code and datasets of this work will be released at https://github.com/opendatalab/MLS-BRN.git.