@Prakash Yogi I would recommend trying the QuickStart to use the API using any of the languages. This quickstart should help you detect and identify faces from person groups by creating one. The concept of person groups & directory structure essentially helps you manage an individual's identity where they hold the uploaded person data including face recognition features and, in most cases, customers use the groups to identify users in their application and external databases to handle business logic for their applications. Using blob storage along with the APIs will help you maintain a structure for storing the images as per your business logic and you can name the person groups and blob folders using a matching naming convention for easy retrieval.
The easiest way to try all the concepts without writing any code is to use the API reference along with your resource keys and then try to create an application using any of the available client SDKs. Thanks!!