Innovating at the Edge with Open Source Software
summary
Claims concrete: uses video analytics serving (GStreamer + OpenVINO) as a REST-driven pipeline manager, ties to EdgeX Foundry for camera discovery, and uses Flogo + Project Air to compile low‑code flows into Go binaries and containers. Notes that networks start to choke after about 9–10 cameras, that edge clusters are typically one–three nodes, and a real‑time sensor‑fusion loss‑prevention reference implementation is on GitHub and boots from a fresh node to running microservices.
AI-written summary · about
chapters (16)
- 0:00Intro
- 1:10Welcome
- 1:40Edge Computing
- 3:45Continuity of Compute
- 6:23Embedded Computing vs Edge Computing
- 10:12Three Ways to Find the Edge
- 11:58Computer Vision
- 12:57Deep Learning
- 15:31Challenges
- 18:22Fun Projects
- 21:55EdgeX
- 23:38EdgeX Foundry
- 24:09flogo
- 26:40Project Air
- 32:46Conclusion
- 36:11Links
related talks
26mOpen Source and Edge Computing: Opening a New Chapter in Digital Transformation
33mThe Case for Open Source in Autonomous AI Systems in the Edge
19mBootstrapping edge infrastructure for AIoT applications with open source software
28mBuilding a Sustainable Future through Open Source
22m#8 Open Source Forum 2019 Jonas Werner
17mIoT: Boom or Bust?… And Will the Edge Really Eat the Cloud?
17mKeynote: Connecting the Dots on All Things Great and Small
41mOpen Source in Action: Pioneering OSS in Real World Production