Dataset · dronefreak
Annotated traffic lights for object detection in YOLO format
dronefreak/LISA-Traffic-Lights
An unofficial conversion of the LISA dataset with traffic light bounding boxes for training detectors.
What's inside
The set consists of road scene images with traffic light annotations in the form of bounding boxes. This is an unofficial repackaging of the original LISA Traffic Light Dataset by researchers from the Laboratory for Intelligent and Safe Automobiles. The data is exported to YOLO format and distributed under the CC BY-NC-SA 4.0 license.
What you can do with it
- Training YOLO family detectors for traffic light recognition
- Prototyping advanced driver assistance systems (ADAS) and autonomous vehicles
- Validating small object detection models on road scenes
- Tasks
- Object detection
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 20K
- Likes
- 4
Models trained on it
- dronefreak/lisa-yolo26lObject detection 77
- dronefreak/lisa-yolo26nObject detection 68
- dronefreak/lisa-yolo26xObject detection 66
- dronefreak/lisa-yolo26sObject detection 61
- dronefreak/lisa-yolo11xObject detection 59
- dronefreak/lisa-yolov8mObject detection 47
- dronefreak/lisa-yolo26mObject detection 36
- dronefreak/lisa-rfdetr-smallObject detection 22
How to load it
from datasets import load_dataset
ds = load_dataset("dronefreak/LISA-Traffic-Lights")