Skip to content

Latest commit

 

History

History
48 lines (30 loc) · 1.57 KB

README.md

File metadata and controls

48 lines (30 loc) · 1.57 KB

Face_ID

  • AIM

    To create a simple and reliable Face Recognition program (For this VIDEO) by using Tensorflow in the Python Programming Language.

  • Setup

    You will need to install Tensorflow and Python on your Machine. To install tensorflow, hit the following in the terminal:

    pip install tensorflow
    pip install opencv-python
    

    I'm using OpenCV for taking a realtime image of our faces from our WebCamera and then Identify that image from the retrained model! In this way, our program of face_ID will work.

    NOTE: You must have a WebCamera, either on your Laptop or Computer!

    That's enough for (this)[] video. Also, you need to clone this Repository, to do so, hit the following:

    git clone https://github.com/MauryaRitesh/Face_ID
    
  • Gathering Images

    You will need a lot of images of your(or the one for whom you are creating the Face ID) Faces, i.e., go ahead and take hundreds of thousands of selfies and fill all those in the Images folder of this repository. You will need to create a Folder and rename as your name. E.g., Create a folder named Ritesh and another as Not Ritesh.

  • (Re)Training

    To train the model, hit the following:

    python -m scripts.retrain \
    --output_graph=tf_files/retrained_graph.pb \
    --output_labels=tf_files/retrained_labels.txt \
    --architecture=inception_v3 \
    --image_dir=tf_files/images
    
  • Using the Model

    To get the model Working, hit the following:

    python -m scripts.label_image
    

    That's enough for now! Do STAR this Repo if it really helped you!