bendermh/anyeye

NystagmusCNN

ANyEye

Getting started

Python==3.7

1. git clone or download this repository

2. Install requirements on your virtual environment with

    pip install -r requirements.txt
   

3. Edit config.ini

    [config]
    input_dir = <directory of your image or video>
    filetype = <image or video>
    num_eyes = <count eyes of your input>
    left_eye_index = <insert index of left (inverted) eye if exists>
    video_out = <1 if you want video result 0 if you don't>
    model = <unet, unet++l2, unet++l3, or unet++l4>
   

4. Execute ANyEye with

    python anyeye.py
   

5. Results will be created in ./results