WebAssembly-based Edge AI Sensing Pipeline Orches...
summary
A platform that compiles and deploys small WebAssembly modules to far‑edge devices is already built: EVP agent runs on Linux and MCU-class devices, reporting hardware to a cloud sensing pipeline service that does dynamic AOT compilation to produce device‑specific Wasm binaries. Demo shows license‑plate detection/ocr deployed to Raspberry Pi and x86, with modules wired by Node‑RED, ThingsBoard provisioning, and OpenCV image ops; performance is “near native” versus native after AOT. Key components will be open source.
AI-written summary · about
chapters (9)
- 0:00Start
- 3:23Challenges to Solve
- 7:38EVP (Edge Virtualization Platform) Agent
- 12:27Sensing Pipeline Service Overview - High Level
- 15:11Sensing Pipeline Service Overview - Solution Design
- 17:09Sensing Pipeline Service Overview - Solution Instantiation
- 18:45Demo: License Plate Detection and Reading
- 30:33Device Stack (work in progress)
- 32:58Ecosystem Expansion
related talks
07mKeynote: Modernizing the Development of AI IoT Devices with WebAssembly-based...- 07mKeynote: Modernizing the Development of AI IoT Devices with WebAssembly-based...
38mStreamlining AI Solutions Development: S...
13mKeynote: WEdge Project: Edge AI enabled by Wasm
42mSecure and Efficient Sensing Applications with Wasm: Sony...
32mWebAssembly Based AI as a Service on the Edge with Kubernetes
32mA Next-generation IoT Platform for Edge AI Apps Leveraging Se...
29mLife on the Edge