This repository will demostrate how to deploy a offical YOLOv7 pre-trained model with OpenVINO runtime api

  $ pip install -r python/requirements.txt

Please follow the Guides to install OpenVINO and OpenCV

Download YOLOv7 pre-trained weight from YOLOv7

  $ git clone [email protected]:WongKinYiu/yolov7.git
  $ cd yolov7/models
  $ python export.py --weights yolov7.pt

The input image can be found in YOLOv7's repository

 $ python python/image.py -m yolov7.onnx -i data/horse.jpg -d "CPU" -p False

You can also try running the code with Preprocessing API for performance optimization.

 $ python python/image.py -m yolov7.onnx -i data/horse.jpg

• -i = path to image or video source;
• -m = Path to IR .xml or .onnx file;
• -d = Device name, e.g "CPU";
• -p = use preprocessing api, True or False
• -bs = Batch size;
• -n = number of infer requests;
• -g = with/without grid in model, True or False;

Compile the source code

  $ cd cpp
  $ mkdir build && cd build
  $ source '~/intel/openvino_2022.1.0.643/bin/setupvars.sh'
  $ cmake ..
  $ make

You can also uncomment the code in CMakeLists.txt to trigger Preprocessing API for performance optimization.

Run inference

 $ yolov7 yolov7.onnx data/horses.jpg 'CPU'

You can also run the sample with webcam for real-time detection

$ python python/webcam.py -m yolov7.onnx -i 0

Tips: you can switch the device name to "GPU" to improve the performance.

Try this notebook (yolov7-optimization) and quantize your YOLOv7 model to INT8.