Train Media Server to Recognize Objects

You must train Media Server by providing images of objects that you want to recognize. When you run object recognition, Media Server uses the training data to recognize objects in your media.

The following diagram shows how Media Server stores information you provide during training.

The "Media Server database" represents the Media Server datastore file or a database on an external database server. For information on setting up this database, see Introduction.

You can organize objects into databases. These object databases are for organizational purposes. For example, you might have a database for company logos, and a database for your company's products. If you want to recognize only one category of objects in your images, you can use only that database.

A database can contain any number of objects, both 2D and 3D. You can associate training options and custom metadata with each object.

To each object you must add a training image (multiple training images for a 3-D object). For information about choosing suitable training images, see Select Images for Training.