The Rise of No-Code AI Vision at the Edge
summary
Shows a working retail item detector built by pairing Node-RED flows with Docker containers on a Jetson Nano class device, using a camera input, a data-loader container and a detection container. Training uses Ultralytics YOLOv5 then converts to a YOLOv5 TensorRT .engine model for on-device inference. Demo uses JetPack 4.6.1, Docker Compose and node packages, cites the Node-RED image’s ~100 million Docker Hub pulls, and boots up and detects items in real time.
AI-written summary · about
chapters (10)
- 0:00Start
- 0:22Speakers
- 3:14When No-Code Meets AI
- 4:12Overview of Node-RED
- 6:43100M+ Docker Hub Pulls for Node-RED
- 9:41Architecture
- 11:41Setup
- 14:07Node-RED Flow Editor
- 14:34Docker Compose
- 29:52References
