![]() ![]() ![]() After end users sign in to Okta, they can launch any of their assigned app integrations to access external applications and services without reentering their credentials. Add the following snippet inside the else block from the code above.The Okta app integrations in your org use Single Sign-On (SSO) to provide a seamless authentication experience for end users. We want only to allow the image to update when we have some measure of confidence that the new face image matches the baseline. Now we want to handle the second scenario - where we want to perform facial analysis compared to the baseline facial features. We're using a simple case here to showcase how this service works by a single image of an individual. We're also associating one face to a PersonGroupPerson, although you can add multiple images of an individual so you can train the service to analyze an individual better. The PersonGroupPerson object is an individual within the group, so there is usually a 1:n relationship between PersonGroup and PersonGroupPerson. Other options could be group or department IDs. ![]() We're using your Okta user ID as the unique identifier for this group, so each group is truly individualized. The PersonGroup is a container object for a group of people. In the EditProfile(UserProfileViewModel profile) method, replace the line of code that calls await UpdateUserImage() with the code snippet below. We can store facial features within the "OktaProfilePicture" Azure Cognitive Services resource, which we'll do for the first upload scenario. The Face API only stores the extracted facial features for 24 hours by default, so we need to set a baseline that we can refer to beyond the 24-hour window. You're updating your profile picture - facial analysis runs against the face in this image and compares it to the baseline.You're adding a profile picture for the first time - this sets the baseline for the facial features to use in future comparisons.Now, if you try running the app and upload an image with you and your friends, you'll see an error. The RecognitionModel.Recognition04 is the most accurate model currently available, and DetectionModel.Detection01 is a model that avoids detecting small and blurry faces. The Azure section of your will now look like the following. Copy and paste the key and endpoint values from the Azure Keys and Endpoint view. Open the file and add two new fields in the Azure section named SubscriptionKey and FaceClientEndpoint. You will need both the key and the Endpoint. Open "OktaProfilePicture" Face service instance and open Keys and Endpoint. Press Review + create to create the resource. Since this is for demo purposes, I used the "Free F0" pricing tier. Select "OktaBlog" as the Resource group (or a Resource group of your choosing) and name the instance "OktaProfilePicture". Press + Create to open the Create Face view. Open the Cognitive Services Face resource page in the Azure portal. We'll use facial analysis in two different ways - for face detection and face verification.įirst, we need to create the Azure resource and get the access keys. ![]() If you run the app, you can select a picture from local files to add to your profile and see your profile picture on the "Profile" page.įinally, we get to check out Azure Cognitive Services for facial analysis. || □ (ViewData != null) Įnter fullscreen mode Exit fullscreen mode Run the following commands to create the solution, the project, and add the project to the = "User Profile" A Microsoft Azure Account (Azure free account).NET projects, such as Visual Studio, Visual Studio for Mac, VS Code, or JetBrains Rider NET 5.0 runtime and SDK, which includes the. You'll also authenticate with Okta and store user data as custom profile attributes.Īt the end of this post, you'll be able to upload a profile picture in your app and get information about image error conditions, such as when zero or more than one face is detected or when your facial features don't match a new picture. NET MVC application and store user profile pictures in Azure Blob Container Storage. In this article, you will learn how to use the Vision Face API to perform facial analysis in a. Azure Cognitive Services has vision, speech, language, and decision-making services. With Azure Cognitive Services, you can add AI capabilities using pre-trained models, so you don't need machine learning or data science experience. Azure Cognitive Services is a collection of cloud-based AI products from Microsoft Azure to add cognitive intelligence into your applications quickly. ![]()
0 Comments
Leave a Reply. |