The system is made with the idea in mind, that the client is ran through a LabView programme, using the Python Integration Toolkit (http://docs.enthought.com/python-for-LabVIEW/guide/install.html)
To launch the segmentation, in edi_rovam_yolov7/added_code run
python server_detect.py --source 0 --send_to_server
It is set to send the found coordinates to local_host. For a different IP, change it in line
serverSocket.bind(("127.0.0.1",9090));
To receive the coordinates, in edi_rovam_yolov7/added_code run
python client.py --live
If running without a live connection to a camera, omit the --live
python client.py
It is set to receive coordinates from local_host. For a diferrent IP, change it in line
server_ip = "127.0.0.1"
-
Download and install >=3.8 version of Python from https://www.python.org/downloads/ 1.1. Test if it's working. In windows explorer go to the Python directory (C:\Users\user\AppData\Local\Programs\Python) 1.2. In the address bar write cmd to open a terminal. Then write python and press enter.
-
Install PyTorch. In https://pytorch.org/get-started/locally/ set the STABLE, WINDOWS, PIP, PYTHON, CPU version. 2.1. Open the python scripts folder ((C:\Users\user\AppData\Local\Programs\Python\Python3\Scripts) 2.2. In address bar write cmd 2.3. Run the pip command in the terminal
-
Install YOLOv7 3.1. git clone or download this package to your computer 3.2. Go to edi_rovam_yolov7 folder and in address bar write cmd 3.3. Install the requirements
pip install -r requirements.txt
3.3. In case of large red error, install Visual Studio from https://visualstudio.microsoft.com/downloads/
Go to https://pytorch.org/get-started/locally/ and select the specifics for your system, for me
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu
Download the package to your computer
git clone https://github.com/oskarsVismanis/edi_rovam_yolov7.git
To set up your Python virtual environment, configure and run
path/to/edi_rovam_yolov7/setup/create_python_env.bash
By default, this will create a Python virtual environment in ~/python-virtualenvs/edi_rovam_yolov7
To then source this virtual environment, run
source path/to/edi_rovam_yolov7/setup/source_edi_rovam_yolov7.bash
As documented in the above script, we recommend making a bash function in your ~/.bashrc script so you can easily just call edi_rovam_yolov7 from your Terminal to get started.
edi_rovam_yolov7() {
source /path/to/edi_rovam_yolov7/setup/source_edi_rovam_yolov7.bash
}
With the virtual environment activated, in edi_rovam_yolov7 install the requirements
pip3 install -r requirements.txt