To identify Facial expressions using Deep neural networks (Convolutional Neural Networks).
The dataset is used from Kaggle (https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge/data) The code used is from the coursera's project based learning course primarily taught by Snehan Kekre (https://www.coursera.org/learn/facial-expression-recognition-keras/home/info)
The model is trained by 4 convolutional layers followed by 2 fully connected layers and then by the Final output layer. The model weights are stored in model_weights.h5. Opensource code is used to crate a flask app which is run on a basic HTML code which projects the final video. Model uses seaborn,matplotlib,numpy,tensorflow and keras for initial training. Opencv is used for face detection and Flask is used to create HTML GUI.
-
Customer Service The technology can also be used by those within the hospitality industry to deliver a greater level of customer service. For example, facial recognition can allow employees to quickly identify guests satisfaction and sentiments, perhaps before they even check-in, and deliver more personalised greetings and a more tailored service.
-
Research and Information In order to understand customers, improve services and optimise processes, hotels and other companies need to be able to gather feedback and data. Sentiment analysis can be done through emotion detection.