Geppetto's Guide to Living Puppets
summary
Prototype animatronic puppet connects webcam facial tracking to real servos and it works live: vision runs in-browser with the BRFv JavaScript library because face‑mesh missed eyelid closure, it samples about 13 frames to detect blinks, computes eyelid/mouth ratios and emits fetch calls to Particle Cloud webhooks, which drive three servo motors on a Particle Argon to blink left, blink right and open the mouth; video and annotated canvas publish over Vonage Video API and Azure Functions handle glue and auth.
AI-written summary · about
chapters (8)
- 0:00Introduction and overview
- 1:06From text to emojis to puppets
- 3:19Designing the puppet system
- 8:31Building the device side
- 18:08Coding the web tracking
- 19:54Live demo and puppet performance
- 22:04Practicality and future ideas
- 23:10Q&A session
related talks
35m3D virtual puppet remote controlled with smartphone
19mHand gesture-controlled robot through the Internet
18mExploring alternative interactions in JavaScript
30mLook ma, no hands! Multimodal AI-Agents in the browser
36mWear glasses with NodeJS, Puppeteer, and Chrome
29mNatural User Interface using JavaScript
31mwith modern_peripherals: Python and Flask
20mWeb based 3D Camera Simulator for Perceptual Computing