• Search by Module
• Search by Words
• Search Projects

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

Project: Smile-Detector (GitHub Link)

• Smile-Detector-master
  • Procfile
  • siddhantbansal
    • callbacks
      • __pycache__
        • trainingmonitor.cpython-36.pyc
        • __init__.cpython-36.pyc
      • trainingmonitor.py
      • __init__.py
    • datasets
      • __pycache__
        • __init__.cpython-36.pyc
        • simpledatasetloader.cpython-36.pyc
      • simpledatasetloader.py
      • __init__.py
    • nn
      • perceptron.py
      • __pycache__
        • neuralnetwork.cpython-36.pyc
        • __init__.cpython-36.pyc
        • perceptron.cpython-36.pyc
      • neuralnetwork.py
      • __init__.py
      • conv
        • minivggnet.py
        • lenet.py
        • __pycache__
          • minivggnet.cpython-36.pyc
          • __init__.cpython-36.pyc
          • shallownet.cpython-36.pyc
          • lenet.cpython-36.pyc
        • __init__.py
        • shallownet.py
    • preprocessing
      • __pycache__
        • simplepreprocessor.cpython-36.pyc
        • __init__.cpython-36.pyc
        • imagetoarraypreprocessor.cpython-36.pyc
        • simpledatasetloader.cpython-36.pyc
        • aspectawarepreprocessor.cpython-36.pyc
      • aspectawarepreprocessor.py
      • imagetoarraypreprocessor.py
      • __init__.py
      • simplepreprocessor.py
  • LICENSE
  • runtime.txt
  • code.py
  • Templates
    • index.html
  • detect_smile.py
  • train_model.py
  • README.md
  • requirements.txt

Smile-Detector

Using deep learning to detect smiles from a live video feed.

Description

This project was made for detecting smile in a live video using the webcam or a pre-recorded video.

Preview

Requirements

• numpy
• matplotlib
• openCV
• sklearn
• imutils
• TensorFlow/Theano
• Keras
• h5py

You can install all the required libraries by running the following command

pip install requirements.txt

Functionalities

1. Filter to detect face.
2. CNN for traning the model(Using LeNet architecture).

Procedure

1. Gather a dataset of people smiling and not smiling which is tighly cropped around the face.
2. Train the Convolutional Neural Network using train_model.py and your custom data set(the network is designed to handle class imbalance).
3. Run the detect_smile.py to start a live video using your webcam, or you can also supply an optional argument to enter the path to a video.
   

   Credits

   Adrian Rosebrock creator of PyimageSearch