This is a deep learning model for detecting the potholes on roads. The model is using YOLO-v2.
Download the weights from this link : https://drive.google.com/open?id=1VwViHDdc4W8t28rMiuGjxChiAtFxMAjN and place the weights in the root folder.
Then
python predict.py -c config.json -w /path/to/best_weights.h5 -i /path/to/image/or/video
The model has been trained with 150 images having 568 potholes labels. The accuracy of the model will increase by traing the model with larger size dataset.
