Backgrounds are blurry, but the future is clear
summary
Explains how browser ML (TensorFlow WASM, MediaPipe, BodyPix and smaller pre-trained models) enables virtual backgrounds and background blur in WebRTC. Covers how segmentation models work, how to integrate an ML segmentation library (example: MediaPipe) into a WebRTC video call (Daily integration is walked through), and describes other custom video-processing applications you can build on that stack.
AI-written summary · about
related talks
28mThe Doctor Will See You Now: Machine Learning for Telehealth
24mWeb Apps of the Future With Web AI
19mML at Client Side
24mThe next level apps for Web developers Machine learning on browsers 🤯
24mState of client side machine learning
34mThe Browser, The Machine Learning and The Future we might not have imagined just less than a year ago
51mRealTime Exoskeleton-Mapping and Body Segmentation on the Browser with Tensorflow.js
14mVisual Blocks: Visual prototyping of AI pipelines