- Go to the folder project.
$ cd python-object-recognition
- Activate the python environment.
$ source bin/activate
- Run the project.
$ python3 gui.py
- CTRL + O: open file (only on video).
- space: start/stop (video and real time).
- CTRL + R: start/stop recognition (video and real time).
(MacOS users need to use command key instead the CTRL).
- the outputs will always be printed to the console.
- for MacOS users the screen reader voice Over will speak all recognized objects.
- Add new objects (images) to the neural network.
- Train the neural network.
- Add ESpeak screen reader to the project (available voice responses to Linux, MacOS and Windows users).
- Create a virtual environment.
- Activate the virtual environment.
- Install WXPython.
pip3 install wxpython
- Install numpy.
pip install numpy
- Install OpenCV 3.
pip install opencv-python
- Install Tensor flow:
pip3 install --upgrade tensorflow
- Run.
python3 gui.py