Build an AI-powered “green” app with Azure Custom Vision header image

Build an AI-powered “green” app with Azure Custom Vision

On Thursday afternoon, I had a fascinating time presenting at the 29th AzureHeads meetup about how to create image classification models in Azure Custom Vision. I talked about AI and Computer Vision, explained the concept of Transfer Learning, and then showed how Azure Custom Vision works. I hope that everyone enjoyed the session and learned new skills!

The session

Azure Custom Vision is an Azure Cognitive Services service that lets you build and deploy your own image classification and object detection models. In this session, we will explain the concept of AI and custom vision and build an image classification model for agriculture with the Azure Custom Vision service. We will then deploy and consume the model in a Python app and discuss the available options for exporting the Custom Vision model to run offline.

What you will learn in this session:
  • Understand the concept of AI, Computer Vision and Transfer Learning.
  • Use the pretrained machine learning models in Azure Custom Vision.
  • Build and deploy an image classification model and consume the model in a Python app.

Self-paced learning from Microsoft Learn

Sketch note

The following sketch note is a visual guide to my session presented at the AzureHeads meetup and describes the basic features of Azure Custom Vision.

Azure Custom Vision

Learning resources

If you are interested in getting started with Computer Vision in Microsoft Azure, you can read the articles that I have written about Azure Custom Vision.

Build a flower classification model with Azure Custom Vision header image
Build a flower classification model with Azure Custom Vision

In this article, you will build and deploy an image classification model for flower classification with the Azure Custom Vision service.

Featured on the Educator Developer Blog – Microsoft Tech Community

Related Posts

Begin typing your search term above and press enter to search. Press ESC to cancel.

Back To Top