Mapping Rio de Janeiro's favelas: general-purpose vs. satellite-specific neural networks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918154456793088 |
|---|---|
| author | Hallopeau, Thomas Guérin, Joris Demagistri, Laurent Fouzai, Youssef Gracie, Renata De Matos, Vanderlei Pascoal Gurgel, Helen Dessay, Nadine |
| author_facet | Hallopeau, Thomas Guérin, Joris Demagistri, Laurent Fouzai, Youssef Gracie, Renata De Matos, Vanderlei Pascoal Gurgel, Helen Dessay, Nadine |
| contents | While deep learning methods for detecting informal settlements have already been developed, they have not yet fully utilized the potential offered by recent pretrained neural networks. We compare two types of pretrained neural networks for detecting the favelas of Rio de Janeiro: 1. Generic networks pretrained on large diverse datasets of unspecific images, 2. A specialized network pretrained on satellite imagery. While the latter is more specific to the target task, the former has been pretrained on significantly more images. Hence, this research investigates whether task specificity or data volume yields superior performance in urban informal settlement detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_03725 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mapping Rio de Janeiro's favelas: general-purpose vs. satellite-specific neural networks Hallopeau, Thomas Guérin, Joris Demagistri, Laurent Fouzai, Youssef Gracie, Renata De Matos, Vanderlei Pascoal Gurgel, Helen Dessay, Nadine Computer Vision and Pattern Recognition Machine Learning While deep learning methods for detecting informal settlements have already been developed, they have not yet fully utilized the potential offered by recent pretrained neural networks. We compare two types of pretrained neural networks for detecting the favelas of Rio de Janeiro: 1. Generic networks pretrained on large diverse datasets of unspecific images, 2. A specialized network pretrained on satellite imagery. While the latter is more specific to the target task, the former has been pretrained on significantly more images. Hence, this research investigates whether task specificity or data volume yields superior performance in urban informal settlement detection. |
| title | Mapping Rio de Janeiro's favelas: general-purpose vs. satellite-specific neural networks |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2510.03725 |