# Learning from Unlabelled Data with Transformers: Domain Adaptation for Semantic Segmentation of High Resolution Aerial Images

arXiv: https://arxiv.org/abs/2404.11299 · HTML: https://mlsanity.com/paper/2404.11299/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 34 references that are in the map.
- First posted on arXiv 17 Apr 2024.

## Builds on (most cited first, 21 of 21)

- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence](https://mlsanity.com/paper/2001.07685/) (2020)
- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [Segment Anything in High Quality](https://mlsanity.com/paper/2306.01567/) (2023)
- [Fast Segment Anything](https://mlsanity.com/paper/2306.12156/) (2023)
- [Wide-Area Image Geolocalization with Aerial Reference Imagery](https://mlsanity.com/paper/1510.03743/) (2015)
- [Semantic-SAM: Segment and Recognize Anything at Any Granularity](https://mlsanity.com/paper/2307.04767/) (2023)
- [UNetFormer: A UNet-like Transformer for Efficient Semantic Segmentation of Remote Sensing Urban Scene Imagery](https://mlsanity.com/paper/2109.08937/) (2021)
- [Where am I looking at? Joint Location and Orientation Estimation by Cross-View Matching](https://mlsanity.com/paper/2005.03860/) (2020)
- [Domain-Symmetric Networks for Adversarial Domain Adaptation](https://mlsanity.com/paper/1904.04663/) (2019)
- [Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives](https://mlsanity.com/paper/2208.07422/) (2022)
- [Robustness of SAM: Segment Anything Under Corruptions and Beyond](https://mlsanity.com/paper/2306.07713/) (2023)
- [Semi-Supervised Semantic Segmentation in Earth Observation: The MiniFrance Suite, Dataset Analysis and Multi-task Network Study](https://mlsanity.com/paper/2010.07830/) (2020)
- [AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation](https://mlsanity.com/paper/2306.06842/) (2023)
- [DIAL: Deep Interactive and Active Learning for Semantic Segmentation in Remote Sensing](https://mlsanity.com/paper/2201.01047/) (2022)
- [MSMatch: Semi-Supervised Multispectral Scene Classification with Few Labels](https://mlsanity.com/paper/2103.10368/) (2021)
- [GeoMultiTaskNet: remote sensing unsupervised domain adaptation using geographical coordinates](https://mlsanity.com/paper/2304.07750/) (2023)
- [An Interpretable Deep Semantic Segmentation Method for Earth Observation](https://mlsanity.com/paper/2210.12820/) (2022)

## Built on by (most cited first, 4 of 4)

- [Evaluating and Benchmarking Foundation Models for Earth Observation and Geospatial AI](https://mlsanity.com/paper/2406.18295/) (2024)
- [Generalizable Multispectral Land Cover Classification via Frequency-Aware Mixture of Low-Rank Token Experts](https://mlsanity.com/paper/2505.14088/) (2025)
- [Building Age Estimation: A New Multi-Modal Benchmark Dataset and Community Challenge](https://mlsanity.com/paper/2502.13818/) (2025)
- [Enhancing Ground-to-Aerial Image Matching for Visual Misinformation Detection Using Semantic Segmentation](https://mlsanity.com/paper/2502.06288/) (2025)

## Abstract (arXiv)

Data from satellites or aerial vehicles are most of the times unlabelled. Annotating such data accurately is difficult, requires expertise, and is costly in terms of time. Even if Earth Observation (EO) data were correctly labelled, labels might change over time. Learning from unlabelled data within a semi-supervised learning framework for segmentation of aerial images is challenging. In this paper, we develop a new model for semantic segmentation of unlabelled images, the Non-annotated Earth Observation Semantic Segmentation (NEOS) model. NEOS performs domain adaptation as the target domain does not have ground truth semantic segmentation masks. The distribution inconsistencies between the target and source domains are due to differences in acquisition scenes, environment conditions, sensors, and times. Our model aligns the learned representations of the different domains to make them coincide. The evaluation results show that NEOS is successful and outperforms other models for semantic segmentation of unlabelled data.