VisDrone2YOLO Convert annotations from the VisDrone dataset to the YOLO format Original VisDrone classes are converted to the following: 0: pedestrian 1: people 2: bicycle 3: car 4: van 5: truck 6: tricycle 7: awning-tricycle 8: bus 9: motor