Labeled Dataset for Object Detection in Carla Simulator
This dataset contains 1028 images each 640x380 pixel. The dataset is split into 249 test and 779 training examples.
Every image comes with an associated label .xml file in the pascal VOC format (labels folder), in the yolo format (labels_yolo_format) as well as annotations in the MS COCO format (annotations folder) as a json file. The dataset was collected in Carla Simulator, driving around in autopilot mode in various environments (Town01, Town02, Town03, Town04, Town05) and saving every x-th frame. The labels where then automatically generated using the semantic segmentation information.
Please have a look at my other repository to train your custom yolov7 on this CARLA Object Detection Dataset: https://github.com/DanielHfnr/yolov7-carla-object-detection
Available classes are:
- Vehicle (Car, Truck)
- Bike
- Motobike
- Traffic light
- Traffic sign
Example image:
- Restructured dataset
- Changed train/val/test split
- Changed folder structure
- Added annotations in MS COCO format