Real time human facial expression detection

• Main Author: Chong, Grace Kai Xin
• Other Authors: Wang Han
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/149372
• Institution: Nanyang Technological University

Human Facial Expression detection is an important scope in research with regards to human computer interaction. In this project, Convolutional Neural Networks (CNN) is utilized to detect facial expressions, where the model should be able to identify seven emotions (happy, sad, disgust, angry, surprised, fear, neutral) based on an input image or a live webcam feed. CNN of different depths were trained using grayscale images from the Kaggle website using Tensorflow and Keras. Using different networks while tuning different hyperparameters were explored and their effects on the accuracy of predicting the correct output. State-of-the-Art models were also taken inspiration from and used for this task.