# Machine-learned 3D Building Vectorization from Satellite Imagery

arXiv: https://arxiv.org/abs/2104.06485 · HTML: https://mlsanity.com/paper/2104.06485/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 38 references that are in the map.
- First posted on arXiv 13 Apr 2021.

## Builds on (most cited first, 13 of 13)

- [Image-to-Image Translation with Conditional Adversarial Networks](https://mlsanity.com/paper/1611.07004/) (2016)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Conditional Generative Adversarial Nets](https://mlsanity.com/paper/1411.1784/) (2014)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks](https://mlsanity.com/paper/1809.00219/) (2018)
- [Least Squares Generative Adversarial Networks](https://mlsanity.com/paper/1611.04076/) (2016)
- [Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundaries](https://mlsanity.com/paper/1803.08673/) (2018)
- [Evaluation of CNN-based Single-Image Depth Estimation Methods](https://mlsanity.com/paper/1805.01328/) (2018)
- [Auxiliary Tasks in Multi-task Learning](https://mlsanity.com/paper/1805.06334/) (2018)
- [Machine-learned Regularization and Polygonization of Building Segmentation Masks](https://mlsanity.com/paper/2007.12587/) (2020)
- [Late or Earlier Information Fusion from Depth and Spectral Data? Large-Scale Digital Surface Model Refinement by Hybrid-cGAN](https://mlsanity.com/paper/1904.09935/) (2019)
- [ResDepth: Learned Residual Stereo Reconstruction](https://mlsanity.com/paper/2001.08026/) (2020)
- [A Generalized Multi-Task Learning Approach to Stereo DSM Filtering in Urban Areas](https://mlsanity.com/paper/2004.02493/) (2020)

## Built on by (most cited first, 9 of 9)

- [ResDepth: A Deep Residual Prior For 3D Reconstruction From High-resolution Satellite Images](https://mlsanity.com/paper/2106.08107/) (2021)
- [ImpliCity: City Modeling from Satellite Images with Deep Implicit Occupancy Fields](https://mlsanity.com/paper/2201.09968/) (2022)
- [HTC-DC Net: Monocular Height Estimation from Single Remote Sensing Images](https://mlsanity.com/paper/2309.16486/) (2023)
- [GLObal Building heights for Urban Studies (UT-GLOBUS) for city- and street- scale urban simulations: Development and first applications](https://mlsanity.com/paper/2205.12224/) (2022)
- [GroundUp: Rapid Sketch-Based 3D City Massing](https://mlsanity.com/paper/2407.12739/) (2024)
- [Real-GDSR: Real-World Guided DSM Super-Resolution via Edge-Enhancing Residual Network](https://mlsanity.com/paper/2404.03930/) (2024)
- [Sat2lod2: A Software For Automated Lod-2 Modeling From Satellite-Derived Orthophoto And Digital Surface Model](https://mlsanity.com/paper/2204.04139/) (2022)
- [Polygonizing Roof Segments from High-Resolution Aerial Images Using Yolov8-Based Edge Detection](https://mlsanity.com/paper/2503.09187/) (2025)
- [SwiftGS: Episodic Priors for Immediate Satellite Surface Recovery](https://mlsanity.com/paper/2603.18634/) (2026)

## Abstract (arXiv)

We propose a machine learning based approach for automatic 3D building reconstruction and vectorization. Taking a single-channel photogrammetric digital surface model (DSM) and panchromatic (PAN) image as input, we first filter out non-building objects and refine the building shapes of input DSM with a conditional generative adversarial network (cGAN). The refined DSM and the input PAN image are then used through a semantic segmentation network to detect edges and corners of building roofs. Later, a set of vectorization algorithms are proposed to build roof polygons. Finally, the height information from the refined DSM is added to the polygons to obtain a fully vectorized level of detail (LoD)-2 building model. We verify the effectiveness of our method on large-scale satellite images, where we obtain state-of-the-art performance.