atulapra/Emotion-detection
A deep convolutional neural network that classifies facial expressions into seven emotion categories from grayscale 48x48 face images.
Not currently ranked — collecting fresh signals.
star history
This project trains a CNN on the FER-2013 dataset to detect emotions (angry, disgusted, fearful, happy, neutral, sad, surprised) from face images. It uses OpenCV with Haar cascade classifiers for real-time face detection and TensorFlow/Keras for model training and inference. The model achieves approximately 63% test accuracy after 50 epochs and supports real-time webcam-based emotion detection.
Frequently asked
- What is atulapra/Emotion-detection?
- A deep convolutional neural network that classifies facial expressions into seven emotion categories from grayscale 48x48 face images.
- Is Emotion-detection open source?
- Yes — atulapra/Emotion-detection is open source, released under the MIT license.
- What language is Emotion-detection written in?
- atulapra/Emotion-detection is primarily written in Python.
- How popular is Emotion-detection?
- atulapra/Emotion-detection has 1.4k stars on GitHub.
- Where can I find Emotion-detection?
- atulapra/Emotion-detection is on GitHub at https://github.com/atulapra/Emotion-detection.