https://github.com/architgargpro/emotiondetector

This Machine Learning project aims to classify the emotion on a person's face into one of four categories, using deep convolutional neural networks.
https://github.com/architgargpro/emotiondetector

cnn-keras deep-neural-networks keras-tensorflow machine-learning python

Last synced: 26 days ago
JSON representation

This Machine Learning project aims to classify the emotion on a person's face into one of four categories, using deep convolutional neural networks.

• Host: GitHub
• URL: https://github.com/architgargpro/emotiondetector
• Owner: ArchitGargPro
• Created: 2019-08-25T15:45:20.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-08-25T15:53:50.000Z (over 5 years ago)
• Last Synced: 2024-11-19T05:05:06.803Z (3 months ago)
• Topics: cnn-keras, deep-neural-networks, keras-tensorflow, machine-learning, python
• Language: Python
• Homepage:
• Size: 8.24 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: Readme.md

Awesome Lists containing this project

README

        
# Face and Emotion Detection

## Introduction

This project aims to classify the emotion on a person's face into one of four categories, using deep convolutional neural networks. This repository is an implementation of [this](https://github.com/atulapra/Emotion-detection/blob/master/ResearchPaper.pdf) research paper and some help from [this](https://github.com/atulapra/Emotion-detection) GitHub repo. 

## Dataset

The model is trained on the FER-2013 dataset which was published on International Conference on Machine Learning (ICML) available [here](https://anonfile.com/bdj3tfoeba/data_zip). This dataset consists of 35887 grayscale, 48x48 sized face images with seven emotions - angry, disgusted, fearful, happy, neutral, sad and surprised.

## Dependencies

* Python 3.6

* [OpenCV](https://opencv.org/)

* [Tensorflow](https://www.tensorflow.org/)

* [Keras.](https://keras.io/)

## Usage

* First clone or downlad this repository then enter the folder: 'cd emotionDetector'

* Now, you can either use this pretrained modelby loading the weights or train your oun custom model:

	The code is modified in order to use the pretrained model, to use the model use this command: 'python edModel.py'

	To train your model, uncomment code at line 18-74 and place your dataset in data/train/ and data/test/ directories.

	Then use this command 'python edModel.py --mode train' to start training, then 'python edModel.py --mode display'

## Algorithm

* First, we use **haar cascade** to detect faces in each frame of the webcam feed.

* The region of image containing the face is resized to **48x48** and is passed as input to the ConvNet.

* The network outputs a list of **softmax scores** for the seven classes. (we have modified it to give result out of 4 emotions)

* The emotion with maximum score is displayed on the screen.

## References

  "Challenges in Representation Learning: A report on three machine learning contests." I Goodfellow, D Erhan, PL Carrier, A Courville, M Mirza, B

   Hamner, W Cukierski, Y Tang, DH Lee, Y Zhou, C Ramaiah, F Feng, R Li,  

   X Wang, D Athanasakis, J Shawe-Taylor, M Milakov, J Park, R Ionescu,

   M Popescu, C Grozea, J Bergstra, J Xie, L Romaszko, B Xu, Z Chuang, and

   Y. Bengio. arXiv 2013.