README.md

Gender-Paint

A Deep Learning Model to Predict Gender From Paitings

Model Architecture-

The keras model is created by training SmallerVGGNet from scratch on around 2200 face images (~1100 for each class). Face region is cropped by applying face detection using cvlib on the images gathered from Google Images. It acheived around 96% training accuracy and ~90% validation accuracy. (20% of the dataset is used for validation)

Screenshots

Sample Input

Sample Output

Technologies used

• Tensorflow
• Keras
• Flask
• OpenCV

Steps to run user interface

1. Run the file Gender Detection.exe (Detailed front end source code can be found here: https://github.com/upam00/Gender-Paint/tree/Front-End)

Steps to run the application

1. cd Gender-Paint

2. pip3 install -r requirements.txt (only for the first time) then python3 app.py

3. Add or Edit the routes and functions in the file app.py

Testing the API

1. Locally, eg: http://localhost:3434/status
2. With LIVE Heroku Server, eg: https://genderpaintapi.herokuapp.com/status
3. Test the API with POSTMAN.

Example input for POST:

• Set the URL TO https://genderpaintapi.herokuapp.com//upload

INPUT:

    {
        "base64":"Add Your Base64 string here"
    }

---

Authors

Adittya Dey

Subhasish Goswami

Upam Sarmah