Official Installation of YOLOv5
The original code from yolov5 can be easily used for object detection. One can conveniently install your yolov5 following the link above.
This repository offers a simple function that could be used for detecting any single image instead of images in common datasets.
Also with this function, you can get the result of the coordinates of detection b-boxes.
single_img_detect.pyThis file is the main file, modified from the original file detect.py. The function named run could be used directly instead of using the command line as well as the argparse. So you can run this .py file and get the output results.
I wrote a function named my_load_single_img to process any single image in datasets.py, which is imported in single_img_detect.py. You need to replace this datasets.py in the folder of the official code (the link above), utils/datasets.py.
-
Run the
single_img_detect.py, you may get the output image as below.
