Try this demo for an object detector trained on RSUD20K, introduced in our paper RSUD20K: A Dataset for Road Scene Understanding In Autonomous Driving.
RSUD20K is a new object detection dataset for road scene understanding, comprised of over 20K high-resolution images from the driving perspective on Bangladesh roads, and includes 130K bounding box annotations for 13 objects.
You can use this demo to get the a bounding box and class label predictions of objects present in your images or videos. To use it, simply upload an image or video of your choice and hit submit. You will get one or more bounding boxes and names of objects present in your images or videos from this list: (person, rickshaw, rickshaw van, auto rickshaw, truck, pickup truck, private car, motorcycle, bicycle, bus, micro bus, covered van, human hauler)
Dataset Page