This project utilizes the ml5.js library to build a machine learning model that can detect whether someone is wearing a mask in a Twilio Video call. The model uses a pre-trained MobileNet model and Feature Extractor to recognize images of people wearing masks or not. The user can add training data by taking pictures with their webcam, train the model, and then use it to predict whether someone is wearing a mask in real-time during a video call. The project showcases how Twilio's Serverless Toolkit can be used to deploy web apps quickly, including video chat applications, and how machine learning can be integrated into these applications for features like mask detection.