In coordination with augmented reality (AR), object recognition using artificial intelligence (AI) and machine learning (ML) have the potential of creating a new way of interacting with the world, particularly in the field of education. A mobile device’s camera can view a scene, and AI algorithms can identify objects such as trees or chairs within the scene without user interaction. Using an automated machine learning (AutoML) tool, our research team trained a model and utilized that model in a real time, in conjunction with an AR application to help chemistry students understand molecular structures. The input to the application is an image, hand-drawn or computer generated. The app runs the image through our trained model, labels it, and uses that label to generate an interactive 3D model of the molecule. Details of the development of the ML model and its connection to an previously developed interactive, AR-based molecule viewing app will be presented. Additionally, implications of using the technology in the chemistry classroom will be presented.

Molecules and machinelearning(ML): Images classified by ML used as input to anaugmented reality app for chemistry education

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.