Object-detection

project for object detection

Before running

Download from Kaggle yolov3 model: https://www.kaggle.com/datasets/valentynsichkar/yolo-coco-data?resource=download

mkdirs models/yolov3

put models and classes into the directory

Build

Launch script settings.sh to configure requirements and dependencies

Command line

Default it use the yolov3 model

$ python main.py --input <path_input> 

if it needs is possibile to use specific model with relative classes

$ python3 main.py --input <path_input> --model <model_input> --class <class_input> 

Function

• search : add string or index class that want to find during object detection
• input : Can be folder, single image or a video file
• create : create a video file with detected class

TODO

• Set virtualenv