Dataset · dronefreak
Marine drone imagery for object detection
dronefreak/SeaDronesSee
An unofficial YOLO-format build of the SeaDronesSee v2 dataset for water object detection.
What's inside
It contains images and text annotations for object detection tasks in English, with a size ranging from 10,000 to 100,000 examples. This is an unofficial reissue of the original University of Tübingen dataset. Data is reformatted into a standard directory structure compatible with YOLO.
What you can do with it
- Training and fine-tuning YOLO family models for maritime object detection
- Developing computer vision systems for search and rescue UAVs
- Tasks
- Object detection
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 28.6K
- Likes
- 3
Models trained on it
- dronefreak/seadronessee-yolov8sObject detection 84
- dronefreak/seadronessee-yolo11xObject detection 68
- dronefreak/seadronessee-yolo26mObject detection 60
- dronefreak/seadronessee-yolo11nObject detection 58
- dronefreak/seadronessee-yolo26sObject detection 54
- dronefreak/seadronessee-rfdetr-mediumObject detection 37
- dronefreak/seadronessee-yolov8mObject detection 34
- dronefreak/seadronessee-rfdetr-smallObject detection 27
How to load it
from datasets import load_dataset
ds = load_dataset("dronefreak/SeaDronesSee")