From the course: Microsoft Azure AI Engineer Associate (AI-102) Cert Prep by Microsoft Press

Unlock this course with a free trial

Join today to access over 25,100 courses taught by industry experts.

Use Azure AI Vision for object detection and analysis

Use Azure AI Vision for object detection and analysis

- [Instructor] My opinion is that, on this part of AI-102 exam, Microsoft wants us to differentiate between the Computer Vision Service, now called Azure AI Vision, versus the custom vision that allows you to use not just the off the shelf machine learning model that Microsoft has trained. I don't mean that to sound off the shelf, to minimize it. All these cognitive services we have to remember are still maintained actively. They're finely-tuned ML models, but Azure AI Vision off the shelf means that we're using the built-in metadata service. We're looking for object detection, scene understanding, image tagging. Some real-world use cases that you might see on your exam could involve Contoso Outdoors tagging photos for product discovery, Fabrikam scanning product shots for inventory accuracy, Woodgrove Travel detecting scenery types for trip albums, or Northwind Medical using AI Vision to detect anomalies for X-ray surgery. So we remember content safety deals with moderation, hate…

Contents