A Python wrapper on pjreddie's implementation (authors' implementation) of YOLO V3 Object Detector on Darknet. This wrapper is also compatible with other Darknet object detection models.

Image source: http://absfreepic.com/free-photos/download/crowded-cars-on-street-4032x2272_48736.html

Prerequisites

• Python 3.6+
• Linux x86-64 Operating System
• NVIDIA CUDA SDK (for GPU version only. Make sure nvcc is available in PATH variable.)

Sample Usage

Note: This sample code requires OpenCV with python bindings installed. (pip3 install opencv-python==3.4.0)

1. Create a directory to host sample code and navigate to it.
2. Download and execute this script to download model files.
3. Create sampleApp.py with following code. Specify SAMPLE_INPUT_IMAGE.

   from pydarknet import Detector, Image
   import cv2
   
   net = Detector(bytes("cfg/yolov3.cfg", encoding="utf-8"), bytes("weights/yolov3.weights", encoding="utf-8"), 0, bytes("cfg/coco.data",encoding="utf-8"))
   
   img = cv2.imread('SAMPLE_INPUT_IMAGE')
   img_darknet = Image(img)
   
   results = net.detect(img_darknet)
       
   for category, score, bounds in results:
       x, y, w, h = bounds
       cv2.rectangle(img, (int(x - w / 2), int(y - h / 2)), (int(x + w / 2), int(y + h / 2)), (255, 0, 0), thickness=2)
       cv2.putText(img, category ,(int(x),int(y)),cv2.FONT_HERSHEY_COMPLEX,1,(255,255,0))
   
   cv2.imshow("output", img)
   cv2.waitKey(0)
   

4. Execute sampleApp.py python sampleApp.py.

Installation

yolo34py comes in 2 variants, CPU Only Version and GPU Version. Installation may take a while since it involves downloading and compiling darknet.

CPU Only Version

This version is configured on darknet compiled with flag GPU = 0.

pip3 install requests # Used to download darknet
pip3 install cython
pip3 install numpy
pip3 install yolo34py

GPU Version:

This version is configured on darknet compiled with flag GPU = 1.

pip3 install requests # Used to download darknet
pip3 install cython
pip3 install numpy
pip3 install yolo34py-gpu

More Information

• For more details on yolo34py (This python wrapper):
  • GitHub: https://github.com/madhawav/YOLO3-4-Py
  • This is the place to discuss issues of yolo34py.
  • Your contributions are greatly appreciated.
• For more details on YOLO V3:
  • Website from Authors: https://pjreddie.com/yolo
• For more details on Darknet, the base API wrapped by this library
  • Website: https://pjreddie.com/darknet/
  • GitHub: https://github.com/pjreddie/darknet

License

• yolo34py (this wrapper) is under Apache License 2.0.
• The version of darknet wrapped by yolo34py is in public domain.