Skip to content

Latest commit

 

History

History
30 lines (29 loc) · 2.55 KB

File metadata and controls

30 lines (29 loc) · 2.55 KB

REAL TIME FACIAL EMOTION RECOGNIZER

Real-time facial emotion recognition is a technology that uses computer vision and deep learning to analyze a person's facial expressions in real-time and determine their emotional state.

Overview

The Real-Time Facial Emotion Recognition System is a Python-based project that utilizes computer vision and deep learning techniques to perform real-time emotion detection from live video streams captured by a camera source. A CNN model is trained to recognize a range of 7 emotions, including 'Angry', 'Disgust', 'Fear', 'Happy', 'Neutral', 'Sad', and 'Surprise'.

Workflow

  1. It all starts with training a CNN model.
  2. The dataset used to train and test the model is DATASET.
  3. You can downlaod the dataset or directly use the dataset in KAGGLE if you want to perform any changes in real-time-facial-emotion-classification-cnn-using-keras.ipynb file.
  4. The trained model generated will be in .h5 format. Here the emotion detection model is model.h5 file.
  5. This model attained 72% training accuracy and 60% validation accuracy. 3eb93191-ea63-4715-b6d2-dcefdb42d0b7
  6. The system employs the Haar Cascade Classifier, loaded from the haarcascade_frontalface_default.xml file, to detect faces in the video frames.
  7. The code continuously captures video frames from the camera source, making real-time processing possible.
  8. The detected faces are passed through the emotion detection model to classify the emotion, and the corresponding emotion label is overlaid on the video frame.
  9. Here is the sample output. For all emotion outputs you can check out OutputScreenshots folder. Screenshot (273)

Prerequisites

  • Python
  • TensorFlow
  • Keras
  • OpenCV
  • Pandas
  • Numpy
  • Seaborn
  • Matplotlib

Usage

  1. You can start by cloning the project repository to your local system or by downloading the zip file and extracting it in your working folder.
  2. Ensure you have the pre-trained emotion detection model (model.h5) and the Haar Cascade Classifier XML file (haarcascade_frontalface_default.xml) placed in the project directory. These models are essential for the system to function.
  3. Execute the Python script main.py to start the real-time facial emotion recognition system.