download model and video demo in https://drive.google.com/drive/folders/1YQXKm2FvZ_tNQOE-HELmSuz33oO3ocyj?usp=sharing
put model vehicle.weight at folder vehicle/weights
put model yolov5*.pt at folder yolov5/weights
-
you can use your detection algorithm with:
output detect must (input image): [[xmin1,ymin1,xmax1,ymax1],[xmin2,ymin2,xmax2,ymax2]....],imgs with imgs = [img1,img2...] img1 = image[ymin1:ymax1,xmin1:xmax1] ...
or my detection algorithm that i proviced:
1) yolov4 :
you can modify config at vehicle/config.py
and modify label object at line 34 vehicle/detect.py from label coco ( 0:person, 1 :car ....)
2) yolov5:
you can modify config at yolov5/config.py
and modify label object at line 39 vehicle/detect.py from label coco ( 0:person, 1 :car ....)
run : put video demo and edit link video at file test.py
if you have any problem please contact me https://www.facebook.com/hoangquoc.hao