How to Boost Performance on an Edge Device

In this tutorial, we will show you the steps needed to boost the performance of your edge device. You will need a hardware accelerator that is supported by alwaysAI - like Intel’s Neural Compute Stick 2. We'll have more information on hardware accelerators in our documentation which will be publicly available soon. Until then, apply for our private beta to get access to our docs.

alwaysAI Demo 3 - Boosting Performance with an Accelerator


1. Begin with a real-time detector starter app and model set.

For this example, we will use the alwaysAI real-time object detector starter app, which uses the MobileNet SSD model.

Here, our object detector app is detecting a potted plant in our office. Note the inference time of the device without the accelerator is .710 seconds.

Don't have access to our object detector starter app? Apply now for our private Beta program.

2. Change the object detection engine

To use the accelerator, change the DNN to DNN Open Vino.


3. Re-deploy and run the start command

Then re-deploy the app using alwaysai app deploy and re-start it using the alwaysai app start command.



4. Check your new inference time

Double-check the inference time from your object detector app in the Streamer on your front-end browser.


You can see after using Intel’s Neural Compute Stick 2 in this example that the inference time has dropped to .094 seconds. That's a significant improvement for this edge device.

We are providing developers with a simple and easy-to-use platform to build and deploy computer vision applications on embedded devices. The alwaysAI Private Beta program is now accepting applications. Join us apply for the beta now.