Practical Examples of TensorFlow.JS, Bringing ML To The Browser

face-api.js — JavaScript API for Face Recognition in the Browser with tensorflow.js

☞ https://t.co/jgbR8xxcjY #DeepLearning #TensorFlow pic.twitter.com/vZaUg9Pp8Y

— Developer Course (@develop_course) July 1, 2019

Pose Detection in the Browser: PoseNet Model: Once upon a time, you needed a device like the Kinect to track the parts of your body. Using TensorFlow.JS and PoseNet, we can now estimate your pose using a web camera. Make sure to check out their demo and work here:
https://github.com/tensorflow/tfjs-models/tree/master/posenet

Web apps built with Magenta.js: During graduate school, I enjoyed doing research in generative art and music using neural networks and an evolutionary strategy called NEAT. So, I especially enjoy seeing how TensorFlow has been applied to music domains. For those that enjoy classical music, you will appreciate this synthetic piano player. For a recurrent neural network, it improvises piano amazingly well. 🙂

Handtrack.js: A library for prototyping realtime handtracking in the browser.