**Subject:** Can Edge Impulse Trained Models Be Deployed in a Web App?

#1

Hi everyone,

I have a quick question. I’m a data science student at a local university in Zimbabwe, currently working on a project idea for a hand sign recognition system. The goal is to develop a web-based application that uses the webcam of my computer to capture and recognize hand signs.

While researching, I came across Edge Impulse, but most of the examples I found focus on deploying models to edge devices, such as tiny microcontrollers.

So my question is:

  • Is it possible to deploy Edge Impulse-trained models in a web app?
    If so, could you provide guidance or point me toward resources that explain how to do this?

Thank you!

#2

Hello @Tawanda,

Yes you can. We have several Linux-based inferencing SDKs that can be ran in web application.

We also have a web assembly (browser) option if you want to run the inference directly in the browser engine (without the need to have a backend, the inference will be ran locally as a web application), or a web assembly (nodeJS) option. Prefer the Linux options above rather than the web assembly (nodeJS) as it won’t leverage hardware acceleration.

Best,

Louis