Poject Structure:
│ ├── yolo-coco │ ├── coco.names │ ├── yolov3.cfg │ └── yolov3.weights │ ├── output │ └── pyolo.py
While running the code you can pass Minimum probability of detection as confidence and non-maxima suppression threshold as threshold. However they have default values of 0.5 and 0.3 respectively.
If you provide an image path as image argument, the result will be saved in output folder. Otherwise the code will use your webcam and its real-time stream.
The main idea is from a tutorial by Adrian Rosebrock and I got to know that thanks to my dear friend Vahidreza
Link to tutorial. link to the repository.
You can also download the weights from this link.
This project is licensed under the Apache License 2.0 - see the LICENSE file for details