Summary of Learning From Unlabelled Data with Transformers: Domain Adaptation For Semantic Segmentation Of High Resolution Aerial Images, by Nikolaos Dionelis et al.
Learning from Unlabelled Data with Transformers: Domain Adaptation for Semantic Segmentation of High Resolution Aerial Images
by Nikolaos Dionelis, Francesco Pro, Luca Maiano, Irene Amerini, Bertrand Le Saux
First submitted to arxiv on: 17 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Non-annotated Earth Observation Semantic Segmentation (NEOS) model successfully addresses the challenges of learning from unlabelled aerial images within a semi-supervised learning framework. By developing a new approach that performs domain adaptation, NEOS aligns the learned representations of different domains to overcome distribution inconsistencies. This allows for accurate semantic segmentation of unlabelled data, outperforming other models in evaluation results. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops a new model called NEOS for semantic segmentation of unlabelled aerial images. This is important because annotating such data accurately can be difficult and costly. The model uses domain adaptation to align the learned representations of different domains. This helps to overcome differences in acquisition scenes, environment conditions, sensors, and times. The results show that NEOS works well and is better than other models at segmenting unlabelled aerial images. |
Keywords
» Artificial intelligence » Domain adaptation » Semantic segmentation » Semi supervised
