AI in React JS
summary
Live demos show client-side ML running in JavaScript with TensorFlow.js and TF Hub models (MobileNet, BlazePose) and a 4 MB NSFW model that hits ~93% accuracy. A reusable AI Lab webcam/local-video React component returns bounding boxes, counts people (30 FPS, ~10 ms draw), filters categories, and triggers actions (alerts, texts, sprinklers). Built apps include enjoyingthe.show (privacy-preserving audience mood), a blight cam, and a squat counter; code is open source and production-ready.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker intro
- 3:36Audience background and AI basics
- 8:10Why AI in JavaScript matters
- 12:43Live demo: Facial emotion detection
- 20:11Introducing AI Lab library
- 22:58Building a security camera with AI
- 27:34Counting people in a video
- 31:25Detecting restricted zones
- 35:15AI Lab features and future
- 38:46Q&A and book giveaway
related talks
46mArtificial Intelligence with JavaScript
30mBuilding Machine Learning Models on Edge with React
11mAI-powered JavaScript applications
25mAI For Everybody
34mAI and JavaScript, no You're Not Dreaming
29mMaking AI better with React feat
02mDiscover Web AI: Client side Agents, Gen AI, and machine learning in the browser
09mAI for React Developers: Opportunities, Learning, and Innovation