We propose a deep-learning-based system for the automatic detection of billboards using consumer-grade unmanned aerial vehicles (UAVs). Thanks to the geospatial information from the device and sensor data, the position of billboards can be estimated.
Note: The dataset is licensed under the
Creative Commons Attribution Non-Commercial 4.0 International. If you want to use the dataset commercially, don't hesitate to contact the authors.
The dataset contains 1361 images supplemented with additional spatial metadata, together with 5210 annotations in a COCO-like format. It can be downloaded from here.
The dataset is also available under the link:
https://doi.org/10.5281/zenodo.8366970
| Object name | Count | Example |
|---|---|---|
| free-standing | 3694 |  |
| wall-mounted | 1284 |  |
| large road sign | 232 |  |
All training scripts and pipelines are located in the ./ml directory. Next instructions are placed in this file.
The application is located in the ./app directory. Tnstallation steps and guide are placed in this file.