# Learning Aerial Image Segmentation from Online Maps

arXiv: https://arxiv.org/abs/1707.06879 · HTML: https://mlsanity.com/paper/1707.06879/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jul 2017.

## Built on by (most cited first, 30 of 30)

- [EarthNets: Empowering AI in Earth Observation](https://mlsanity.com/paper/2210.04936/) (2022)
- [PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation](https://mlsanity.com/paper/2103.06564/) (2021)
- [Mapping oil palm density at country scale: An active learning approach](https://mlsanity.com/paper/2105.11207/) (2021)
- [A diverse large-scale building dataset and a novel plug-and-play domain generalization method for building extraction](https://mlsanity.com/paper/2208.10004/) (2022)
- [AIO2: Online Correction of Object Labels for Deep Learning with Incomplete Annotation in Remote Sensing Image Segmentation](https://mlsanity.com/paper/2403.01641/) (2024)
- [Weakly Supervised Few-Shot Segmentation Via Meta-Learning](https://mlsanity.com/paper/2109.01693/) (2021)
- [Country-wide Retrieval of Forest Structure From Optical and SAR Satellite Imagery With Deep Ensembles](https://mlsanity.com/paper/2111.13154/) (2021)
- [Self-Training Guided Disentangled Adaptation for Cross-Domain Remote Sensing Image Semantic Segmentation](https://mlsanity.com/paper/2301.05526/) (2023)
- [CromSS: Cross-modal pre-training with noisy labels for remote sensing image segmentation](https://mlsanity.com/paper/2405.01217/) (2024)
- [MVPbev: Multi-view Perspective Image Generation from BEV with Test-time Controllability and Generalizability](https://mlsanity.com/paper/2407.19468/) (2024)
- [Task Specific Pretraining with Noisy Labels for Remote Sensing Image Segmentation](https://mlsanity.com/paper/2402.16164/) (2024)
- [Hidden Path Selection Network for Semantic Segmentation of Remote Sensing Images](https://mlsanity.com/paper/2112.05220/) (2021)
- [LOANet: A Lightweight Network Using Object Attention for Extracting Buildings and Roads from UAV Aerial Remote Sensing Images](https://mlsanity.com/paper/2212.08490/) (2022)
- [Deep Learning Based Domain Adaptation Methods in Remote Sensing: A Comprehensive Survey](https://mlsanity.com/paper/2510.15615/) (2025)
- [A hierarchical deep learning framework for the consistent classification of land use objects in geospatial databases](https://mlsanity.com/paper/2104.06991/) (2021)
- [Mastering Spatial Graph Prediction of Road Networks](https://mlsanity.com/paper/2210.00828/) (2022)
- [Slender Object Scene Segmentation in Remote Sensing Image Based on Learnable Morphological Skeleton with Segment Anything Model](https://mlsanity.com/paper/2411.08592/) (2024)
- [On the Adversarial Vulnerabilities of Transfer Learning in Remote Sensing](https://mlsanity.com/paper/2501.11462/) (2025)
- [A Histogram Thresholding Improvement to Mask R-CNN for Scalable Segmentation of New and Old Rural Buildings](https://mlsanity.com/paper/2102.04838/) (2021)
- [3D point cloud segmentation using GIS](https://mlsanity.com/paper/2108.06306/) (2021)
- [AF2: Adaptive Focus Framework for Aerial Imagery Segmentation](https://mlsanity.com/paper/2202.10322/) (2022)
- [Continual Domain Adaptation on Aerial Images under Gradually Degrading Weather](https://mlsanity.com/paper/2308.00924/) (2023)
- [Joint-Optimized Unsupervised Adversarial Domain Adaptation in Remote Sensing Segmentation with Prompted Foundation Model](https://mlsanity.com/paper/2411.05878/) (2024)
- [Copy-Move Forgery Detection and Question Answering for Remote Sensing Image](https://mlsanity.com/paper/2412.02575/) (2024)
- [Parking Space Detection in the City of Granada](https://mlsanity.com/paper/2501.06651/) (2025)
- [Self-Supervised Pretraining for Aerial Road Extraction](https://mlsanity.com/paper/2503.24326/) (2025)
- [Satellite Federated Fine-Tuning for Foundation Models in Space Computing Power Networks](https://mlsanity.com/paper/2504.10403/) (2025)
- [LandSegmenter: Towards a Flexible Foundation Model for Land Use and Land Cover Mapping](https://mlsanity.com/paper/2511.08156/) (2025)
- [Beyond Endpoints: Path-Centric Reasoning for Vectorized Off-Road Network Extraction](https://mlsanity.com/paper/2512.10416/) (2025)
- [Location Prior Generation via Multi-Source Urban Data Fusion for Low-Altitude Air Mobility](https://mlsanity.com/paper/2605.25530/) (2026)

## Abstract (arXiv)

This study deals with semantic segmentation of high-resolution (aerial) images where a semantic class label is assigned to each pixel via supervised classification as a basis for automatic map generation. Recently, deep convolutional neural networks (CNNs) have shown impressive performance and have quickly become the de-facto standard for semantic segmentation, with the added benefit that task-specific feature design is no longer necessary. However, a major downside of deep learning methods is that they are extremely data-hungry, thus aggravating the perennial bottleneck of supervised classification, to obtain enough annotated training data. On the other hand, it has been observed that they are rather robust against noise in the training labels. This opens up the intriguing possibility to avoid annotating huge amounts of training data, and instead train the classifier from existing legacy data or crowd-sourced maps which can exhibit high levels of noise. The question addressed in this paper is: can training with large-scale, publicly available labels replace a substantial part of the manual labeling effort and still achieve sufficient performance? Such data will inevitably contain a significant portion of errors, but in return virtually unlimited quantities of it are available in larger parts of the world. We adapt a state-of-the-art CNN architecture for semantic segmentation of buildings and roads in aerial images, and compare its performance when using different training data sets, ranging from manually labeled, pixel-accurate ground truth of the same city to automatic training data derived from OpenStreetMap data from distant locations. We report our results that indicate that satisfying performance can be obtained with significantly less manual annotation effort, by exploiting noisy large-scale training data.