This software is developed by Jeffrey Poon (Student id: 100363295),
supervised by Dr Jeannette Chin.
TensorFlow.js is a powerful library that brings machine learning capabilities to JavaScript environments, allowing developers to build and execute models directly in web browsers or on Node.js, which is how this web app is made.
This tool enables the integration of machine learning into web applications without relying on server-side processing, making AI more accessible and interactive. When you use this web app, you didn't need to download any machine learning models, and that is
because of the magic of tensorflow.js. Allowing client side services also means that you don't have to worry about compromising privacy, so your web cam input in this web app is completely safe from me :D .
Tensorflow.js has a lot of pretrained ML models, hand guesture recognition in this app is just one of them.
For example, TensorFlow.js can be used to perform real-time image and speech recognition directly in the browser, enhancing user interfaces with AI-driven features like live photo classification or voice commands.
If you want to learn more, please check out the TensorFlow.js website.
As a bonus, below is an extra demo for object classification model using tensorflow, try to show a random object like a keyboard or a cup to the camera, have fun!