https://github.com/rahul1582/emotion-detector

An Emotion Detector Using CNN
https://github.com/rahul1582/emotion-detector

cnn data-augmentation emotion-detector fer2013 opencv

Last synced: 12 days ago
JSON representation

An Emotion Detector Using CNN

• Host: GitHub
• URL: https://github.com/rahul1582/emotion-detector
• Owner: Rahul1582
• License: apache-2.0
• Created: 2020-04-11T07:31:16.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2022-11-21T22:48:34.000Z (almost 2 years ago)
• Last Synced: 2023-03-05T13:52:28.975Z (over 1 year ago)
• Topics: cnn, data-augmentation, emotion-detector, fer2013, opencv
• Language: Python
• Homepage: https://nbviewer.jupyter.org/github/Rahul1582/Emotion-Detector/blob/master/src/emotions.py
• Size: 16.8 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Emotion-detection Using CNN

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

The model is trained on the **FER-2013** dataset.

## Dependencies

To install the required packages, run `pip install -r requirements.txt`.

## Technologies Used

```

1.Python

2.Convolution Neural Network(CNN)

3.Open CV

4.Data Augmentation

```

## To Run

First, clone the repository and enter the folder src

Download the FER-2013 dataset from [here](https://drive.google.com/file/d/1X60B-uR3NtqPd4oosdotpbDgy8KOfUdr/view?usp=sharing) and unzip it inside the `src` folder. This will create the folder `data`.

I had added the dataset into gitignore as it is a very big file.

If you want to train this model, use:  

cd src

python emotions.py --run train

If you want to view the predictions without training again, you can download the pre-trained model from [here](https://drive.google.com/file/d/1FUn0XNOzf-nQV7QjbBPA6-8GLoHNNgv-/view?usp=sharing).

If you want to use the web camera or give input as a video to detect emotions just run-

cd src

python emotions.py --run test

with USE_WEBCAM as True or False

If you want to detect emotions in a picture just run-

cd src

python emotions.py --run picture

With a simple 4-layer CNN, the test accuracy reached 63% in 50 epochs.

 

## Data Preparation (optional)

The [original FER2013 dataset in Kaggle](https://www.kaggle.com/deadskull7/fer2013) is available as a single csv file.

### Demo

Model Summary--