Use Azure AI Custom Vision for object detection

To use the Custom Vision service to create an object detection solution, you need two Custom Vision resources in your Azure subscription:

• An Azure AI Custom Vision training resource – used to train a custom model based on your own training images.
• An Azure AI Custom Vision prediction resource – used to generate predictions from new images based on your trained model.

When you provision the Azure AI Custom Vision service in an Azure subscription, you can choose to create one or both of these resources. This separation of training and prediction provides flexibility. For example, you can use a training resource in one region to train your model using your own image data; and then deploy one or more prediction resources in other regions to support computer vision applications that need to use your model.

Each resource has its own unique endpoint and authentication keys; which are used by client applications to connect and authenticate to the service.

about 2