Edit

Share via

Quickstart: Use the Face service

  • Article

Important

If you are using Microsoft products or services to process Biometric Data, you are responsible for: (i) providing notice to data subjects, including with respect to retention periods and destruction; (ii) obtaining consent from data subjects; and (iii) deleting the Biometric Data, all as appropriate and required under applicable Data Protection Requirements. "Biometric Data" will have the meaning set forth in Article 4 of the GDPR and, if applicable, equivalent terms in other data protection requirements. For related information, see Data and Privacy for Face.

Caution

Face service access is limited based on eligibility and usage criteria in order to support our Responsible AI principles. Face service is only available to Microsoft managed customers and partners. Use the Face Recognition intake form to apply for access. For more information, see the Face limited access page.

Get started with facial recognition using the Face client library for .NET. The Azure AI Face service provides you with access to advanced algorithms for detecting and recognizing human faces in images. Follow these steps to install the package and try out the example code for basic face identification using remote images.

Reference documentation | Library source code | Package (NuGet) | Samples

Prerequisites

  • Azure subscription - Create one for free
  • The Visual Studio IDE or current version of .NET Core.
  • Your Azure account must have a Cognitive Services Contributor role assigned in order for you to agree to the responsible AI terms and create a resource. To get this role assigned to your account, follow the steps in the Assign roles documentation, or contact your administrator.
  • Once you have your Azure subscription, create a Face resource in the Azure portal to get your key and endpoint. After it deploys, select Go to resource.
    • You'll need the key and endpoint from the resource you create to connect your application to the Face API.
    • You can use the free pricing tier (F0) to try the service, and upgrade later to a paid tier for production.

Create environment variables

In this example, write your credentials to environment variables on the local machine that runs the application.

Go to the Azure portal. If the resource you created in the Prerequisites section deployed successfully, select Go to resource under Next Steps. You can find your key and endpoint under Resource Management in the Keys and Endpoint page. Your resource key isn't the same as your Azure subscription ID.

Tip

Don't include the key directly in your code, and never post it publicly. See the Azure AI services security article for more authentication options like Azure Key Vault.

To set the environment variable for your key and endpoint, open a console window and follow the instructions for your operating system and development environment.

  1. To set the VISION_KEY environment variable, replace your-key with one of the keys for your resource.
  2. To set the VISION_ENDPOINT environment variable, replace your-endpoint with the endpoint for your resource.
setx VISION_KEY your-key

setx VISION_ENDPOINT your-endpoint

After you add the environment variables, you may need to restart any running programs that will read the environment variables, including the console window.

Identify and verify faces

  1. Create a new C# application

    Using Visual Studio, create a new .NET Core application.

    Install the client library

    Once you've created a new project, install the client library by right-clicking on the project solution in the Solution Explorer and selecting Manage NuGet Packages. In the package manager that opens select Browse, check Include prerelease, and search for Microsoft.Azure.CognitiveServices.Vision.Face. Select the latest version, and then Install.

  2. Add the following code into the Program.cs file.

    Note

    If you haven't received access to the Face service using the intake form, some of these functions won't work.

    using System;
using System.Collections.Generic;
using System.IO;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;

using Microsoft.Azure.CognitiveServices.Vision.Face;
using Microsoft.Azure.CognitiveServices.Vision.Face.Models;

namespace FaceQuickstart
{
    class Program
    {
        static string personGroupId = Guid.NewGuid().ToString();

        // URL path for the images.
        const string IMAGE_BASE_URL = "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/";

        // From your Face subscription in the Azure portal, get your subscription key and endpoint.
        const string SUBSCRIPTION_KEY = Environment.GetEnvironmentVariable("VISION_KEY");
        const string ENDPOINT = Environment.GetEnvironmentVariable("VISION_ENDPOINT");


         static void Main(string[] args)
        {
            // Recognition model 4 was released in 2021 February.
            // It is recommended since its accuracy is improved
            // on faces wearing masks compared with model 3,
            // and its overall accuracy is improved compared
            // with models 1 and 2.
            const string RECOGNITION_MODEL4 = RecognitionModel.Recognition04;

            // Authenticate.
            IFaceClient client = Authenticate(ENDPOINT, SUBSCRIPTION_KEY);

            // Identify - recognize a face(s) in a person group (a person group is created in this example).
            IdentifyInPersonGroup(client, IMAGE_BASE_URL, RECOGNITION_MODEL4).Wait();

            Console.WriteLine("End of quickstart.");
        }

        /*
         *	AUTHENTICATE
         *	Uses subscription key and region to create a client.
         */
        public static IFaceClient Authenticate(string endpoint, string key)
        {
            return new FaceClient(new ApiKeyServiceClientCredentials(key)) { Endpoint = endpoint };
        }

        // Detect faces from image url for recognition purposes. This is a helper method for other functions in this quickstart.
        // Parameter `returnFaceId` of `DetectWithUrlAsync` must be set to `true` (by default) for recognition purposes.
        // Parameter `FaceAttributes` is set to include the QualityForRecognition attribute. 
        // Recognition model must be set to recognition_03 or recognition_04 as a result.
        // Result faces with insufficient quality for recognition are filtered out. 
        // The field `faceId` in returned `DetectedFace`s will be used in Face - Face - Verify and Face - Identify.
        // It will expire 24 hours after the detection call.
        private static async Task<List<DetectedFace>> DetectFaceRecognize(IFaceClient faceClient, string url, string recognition_model)
        {
            // Detect faces from image URL. Since only recognizing, use the recognition model 1.
            // We use detection model 3 because we are not retrieving attributes.
            IList<DetectedFace> detectedFaces = await faceClient.Face.DetectWithUrlAsync(url, recognitionModel: recognition_model, detectionModel: DetectionModel.Detection03, returnFaceAttributes: new List<FaceAttributeType> { FaceAttributeType.QualityForRecognition });
            List<DetectedFace> sufficientQualityFaces = new List<DetectedFace>();
            foreach (DetectedFace detectedFace in detectedFaces){
                var faceQualityForRecognition = detectedFace.FaceAttributes.QualityForRecognition;
                if (faceQualityForRecognition.HasValue && (faceQualityForRecognition.Value >= QualityForRecognition.Medium)){
                    sufficientQualityFaces.Add(detectedFace);
                }
            }
            Console.WriteLine($"{detectedFaces.Count} face(s) with {sufficientQualityFaces.Count} having sufficient quality for recognition detected from image `{Path.GetFileName(url)}`");

            return sufficientQualityFaces.ToList();
        }

        /*
         * IDENTIFY FACES
         * To identify faces, you need to create and define a person group.
         * The Identify operation takes one or several face IDs from DetectedFace or PersistedFace and a PersonGroup and returns 
         * a list of Person objects that each face might belong to. Returned Person objects are wrapped as Candidate objects, 
         * which have a prediction confidence value.
         */
        public static async Task IdentifyInPersonGroup(IFaceClient client, string url, string recognitionModel)
        {
            Console.WriteLine("========IDENTIFY FACES========");
            Console.WriteLine();

            // Create a dictionary for all your images, grouping similar ones under the same key.
            Dictionary<string, string[]> personDictionary =
                new Dictionary<string, string[]>
                    { { "Family1-Dad", new[] { "Family1-Dad1.jpg", "Family1-Dad2.jpg" } },
                      { "Family1-Mom", new[] { "Family1-Mom1.jpg", "Family1-Mom2.jpg" } },
                      { "Family1-Son", new[] { "Family1-Son1.jpg", "Family1-Son2.jpg" } },
                      { "Family1-Daughter", new[] { "Family1-Daughter1.jpg", "Family1-Daughter2.jpg" } },
                      { "Family2-Lady", new[] { "Family2-Lady1.jpg", "Family2-Lady2.jpg" } },
                      { "Family2-Man", new[] { "Family2-Man1.jpg", "Family2-Man2.jpg" } }
                    };
            // A group photo that includes some of the persons you seek to identify from your dictionary.
            string sourceImageFileName = "identification1.jpg";

            // Create a person group. 
            Console.WriteLine($"Create a person group ({personGroupId}).");
            await client.PersonGroup.CreateAsync(personGroupId, personGroupId, recognitionModel: recognitionModel);
            // The similar faces will be grouped into a single person group person.
            foreach (var groupedFace in personDictionary.Keys)
            {
                // Limit TPS
                await Task.Delay(250);
                Person person = await client.PersonGroupPerson.CreateAsync(personGroupId: personGroupId, name: groupedFace);
                Console.WriteLine($"Create a person group person '{groupedFace}'.");

                // Add face to the person group person.
                foreach (var similarImage in personDictionary[groupedFace])
                {
                    Console.WriteLine($"Check whether image is of sufficient quality for recognition");
                    IList<DetectedFace> detectedFaces1 = await client.Face.DetectWithUrlAsync($"{url}{similarImage}", 
                        recognitionModel: recognitionModel, 
                        detectionModel: DetectionModel.Detection03,
                        returnFaceAttributes: new List<FaceAttributeType> { FaceAttributeType.QualityForRecognition });
                    bool sufficientQuality = true;
                    foreach (var face1 in detectedFaces1)
                    {
                        var faceQualityForRecognition = face1.FaceAttributes.QualityForRecognition;
                        //  Only "high" quality images are recommended for person enrollment
                        if (faceQualityForRecognition.HasValue && (faceQualityForRecognition.Value != QualityForRecognition.High)){
                            sufficientQuality = false;
                            break;
                        }
                    }

                    if (!sufficientQuality){
                        continue;
                    }

                    // add face to the person group
                    Console.WriteLine($"Add face to the person group person({groupedFace}) from image `{similarImage}`");
                    PersistedFace face = await client.PersonGroupPerson.AddFaceFromUrlAsync(personGroupId, person.PersonId,
                        $"{url}{similarImage}", similarImage);
                }
            }

            // Start to train the person group.
            Console.WriteLine();
            Console.WriteLine($"Train person group {personGroupId}.");
            await client.PersonGroup.TrainAsync(personGroupId);

            // Wait until the training is completed.
            while (true)
            {
                await Task.Delay(1000);
                var trainingStatus = await client.PersonGroup.GetTrainingStatusAsync(personGroupId);
                Console.WriteLine($"Training status: {trainingStatus.Status}.");
                if (trainingStatus.Status == TrainingStatusType.Succeeded) { break; }
            }
            Console.WriteLine();

            List<Guid> sourceFaceIds = new List<Guid>();
            // Detect faces from source image url.
            List<DetectedFace> detectedFaces = await DetectFaceRecognize(client, $"{url}{sourceImageFileName}", recognitionModel);

            // Add detected faceId to sourceFaceIds.
            foreach (var detectedFace in detectedFaces) { sourceFaceIds.Add(detectedFace.FaceId.Value); }
            
            // Identify the faces in a person group. 
            var identifyResults = await client.Face.IdentifyAsync(sourceFaceIds, personGroupId);

            foreach (var identifyResult in identifyResults)
            {
                if (identifyResult.Candidates.Count==0) {
                    Console.WriteLine($"No person is identified for the face in: {sourceImageFileName} - {identifyResult.FaceId},");
                    continue;
                }
                Person person = await client.PersonGroupPerson.GetAsync(personGroupId, identifyResult.Candidates[0].PersonId);
                Console.WriteLine($"Person '{person.Name}' is identified for the face in: {sourceImageFileName} - {identifyResult.FaceId}," +
                    $" confidence: {identifyResult.Candidates[0].Confidence}.");

                VerifyResult verifyResult = await client.Face.VerifyFaceToPersonAsync(identifyResult.FaceId, person.PersonId, personGroupId);
                Console.WriteLine($"Verification result: is a match? {verifyResult.IsIdentical}. confidence: {verifyResult.Confidence}");
            }
            Console.WriteLine();
        }
    }
}

  3. Run the application

    Run the application by clicking the Debug button at the top of the IDE window.

Output

========IDENTIFY FACES========

Create a person group (3972c063-71b3-4328-8579-6d190ee76f99).
Create a person group person 'Family1-Dad'.
Add face to the person group person(Family1-Dad) from image `Family1-Dad1.jpg`
Add face to the person group person(Family1-Dad) from image `Family1-Dad2.jpg`
Create a person group person 'Family1-Mom'.
Add face to the person group person(Family1-Mom) from image `Family1-Mom1.jpg`
Add face to the person group person(Family1-Mom) from image `Family1-Mom2.jpg`
Create a person group person 'Family1-Son'.
Add face to the person group person(Family1-Son) from image `Family1-Son1.jpg`
Add face to the person group person(Family1-Son) from image `Family1-Son2.jpg`
Create a person group person 'Family1-Daughter'.
Create a person group person 'Family2-Lady'.
Add face to the person group person(Family2-Lady) from image `Family2-Lady1.jpg`
Add face to the person group person(Family2-Lady) from image `Family2-Lady2.jpg`
Create a person group person 'Family2-Man'.
Add face to the person group person(Family2-Man) from image `Family2-Man1.jpg`
Add face to the person group person(Family2-Man) from image `Family2-Man2.jpg`

Train person group 3972c063-71b3-4328-8579-6d190ee76f99.
Training status: Succeeded.

4 face(s) with 4 having sufficient quality for recognition detected from image `identification1.jpg`
Person 'Family1-Dad' is identified for face in: identification1.jpg - 994bfd7a-0d8f-4fae-a5a6-c524664cbee7, confidence: 0.96725.
Person 'Family1-Mom' is identified for face in: identification1.jpg - 0c9da7b9-a628-429d-97ff-cebe7c638fb5, confidence: 0.96921.
No person is identified for face in: identification1.jpg - a881259c-e811-4f7e-a35e-a453e95ca18f,
Person 'Family1-Son' is identified for face in: identification1.jpg - 53772235-8193-46eb-bdfc-1ebc25ea062e, confidence: 0.92886.

End of quickstart.

Tip

The Face API runs on a set of pre-built models that are static by nature (the model's performance will not regress or improve as the service is run). The results that the model produces might change if Microsoft updates the model's backend without migrating to an entirely new model version. To take advantage of a newer version of a model, you can retrain your PersonGroup, specifying the newer model as a parameter with the same enrollment images.

Clean up resources

If you want to clean up and remove an Azure AI services subscription, you can delete the resource or resource group. Deleting the resource group also deletes any other resources associated with it.

To delete the PersonGroup you created in this quickstart, run the following code in your program:

// At end, delete person groups in both regions (since testing only)
Console.WriteLine("========DELETE PERSON GROUP========");
Console.WriteLine();
DeletePersonGroup(client, personGroupId).Wait();

Define the deletion method with the following code:

/*
 * DELETE PERSON GROUP
 * After this entire example is executed, delete the person group in your Azure account,
 * otherwise you cannot recreate one with the same name (if running example repeatedly).
 */
public static async Task DeletePersonGroup(IFaceClient client, String personGroupId)
{
    await client.PersonGroup.DeleteAsync(personGroupId);
    Console.WriteLine($"Deleted the person group {personGroupId}.");
}

Next steps

In this quickstart, you learned how to use the Face client library for .NET to do basic face identification. Next, learn about the different face detection models and how to specify the right model for your use case.

Specify a face detection model version

Get started with facial recognition using the Face client library for Python. Follow these steps to install the package and try out the example code for basic tasks. The Face service provides you with access to advanced algorithms for detecting and recognizing human faces in images. Follow these steps to install the package and try out the example code for basic face identification using remote images.

Reference documentation | Library source code | Package (PiPy) | Samples

Prerequisites

  • Azure subscription - Create one for free
  • Python 3.x
    • Your Python installation should include pip. You can check if you have pip installed by running pip --version on the command line. Get pip by installing the latest version of Python.
  • Your Azure account must have a Cognitive Services Contributor role assigned in order for you to agree to the responsible AI terms and create a resource. To get this role assigned to your account, follow the steps in the Assign roles documentation, or contact your administrator.
  • Once you have your Azure subscription, create a Face resource in the Azure portal to get your key and endpoint. After it deploys, select Go to resource.
    • You'll need the key and endpoint from the resource you create to connect your application to the Face API.
    • You can use the free pricing tier (F0) to try the service, and upgrade later to a paid tier for production.

Create environment variables

In this example, write your credentials to environment variables on the local machine that runs the application.

Go to the Azure portal. If the resource you created in the Prerequisites section deployed successfully, select Go to resource under Next Steps. You can find your key and endpoint under Resource Management in the Keys and Endpoint page. Your resource key isn't the same as your Azure subscription ID.

Tip

Don't include the key directly in your code, and never post it publicly. See the Azure AI services security article for more authentication options like Azure Key Vault.

To set the environment variable for your key and endpoint, open a console window and follow the instructions for your operating system and development environment.

  1. To set the VISION_KEY environment variable, replace your-key with one of the keys for your resource.
  2. To set the VISION_ENDPOINT environment variable, replace your-endpoint with the endpoint for your resource.
setx VISION_KEY your-key

setx VISION_ENDPOINT your-endpoint

After you add the environment variables, you may need to restart any running programs that will read the environment variables, including the console window.

Identify and verify faces

  1. Install the client library

    After installing Python, you can install the client library with:

    pip install --upgrade azure-cognitiveservices-vision-face

  2. Create a new Python application

    Create a new Python script—quickstart-file.py, for example. Then open it in your preferred editor or IDE and paste in the following code.

    Note

    If you haven't received access to the Face service using the intake form, some of these functions won't work.

    import asyncio
import io
import os
import sys
import time
import uuid
import requests
from urllib.parse import urlparse
from io import BytesIO
# To install this module, run:
# python -m pip install Pillow
from PIL import Image, ImageDraw
from azure.cognitiveservices.vision.face import FaceClient
from msrest.authentication import CognitiveServicesCredentials
from azure.cognitiveservices.vision.face.models import TrainingStatusType, Person, QualityForRecognition


# This key will serve all examples in this document.
KEY = os.environ["VISION_KEY"]

# This endpoint will be used in all examples in this quickstart.
ENDPOINT = os.environ["VISION_ENDPOINT"]

# Base url for the Verify and Facelist/Large Facelist operations
IMAGE_BASE_URL = 'https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/'

# Used in the Person Group Operations and Delete Person Group examples.
# You can call list_person_groups to print a list of preexisting PersonGroups.
# SOURCE_PERSON_GROUP_ID should be all lowercase and alphanumeric. For example, 'mygroupname' (dashes are OK).
PERSON_GROUP_ID = str(uuid.uuid4()) # assign a random ID (or name it anything)

# Used for the Delete Person Group example.
TARGET_PERSON_GROUP_ID = str(uuid.uuid4()) # assign a random ID (or name it anything)

# Create an authenticated FaceClient.
face_client = FaceClient(ENDPOINT, CognitiveServicesCredentials(KEY))

'''
Create the PersonGroup
'''
# Create empty Person Group. Person Group ID must be lower case, alphanumeric, and/or with '-', '_'.
print('Person group:', PERSON_GROUP_ID)
face_client.person_group.create(person_group_id=PERSON_GROUP_ID, name=PERSON_GROUP_ID, recognition_model='recognition_04')

# Define woman friend
woman = face_client.person_group_person.create(PERSON_GROUP_ID, name="Woman")
# Define man friend
man = face_client.person_group_person.create(PERSON_GROUP_ID, name="Man")
# Define child friend
child = face_client.person_group_person.create(PERSON_GROUP_ID, name="Child")

'''
Detect faces and register them to each person
'''
# Find all jpeg images of friends in working directory (TBD pull from web instead)
woman_images = ["https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Mom1.jpg", "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Mom2.jpg"]
man_images = ["https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Dad1.jpg", "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Dad2.jpg"]
child_images = ["https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Son1.jpg", "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Son2.jpg"]

# Add to woman person
for image in woman_images:
    # Check if the image is of sufficent quality for recognition.
    sufficientQuality = True
    detected_faces = face_client.face.detect_with_url(url=image, detection_model='detection_03', recognition_model='recognition_04', return_face_attributes=['qualityForRecognition'])
    for face in detected_faces:
        if face.face_attributes.quality_for_recognition != QualityForRecognition.high:
            sufficientQuality = False
            break
        face_client.person_group_person.add_face_from_url(PERSON_GROUP_ID, woman.person_id, image)
        print("face {} added to person {}".format(face.face_id, woman.person_id))

    if not sufficientQuality: continue

# Add to man person
for image in man_images:
    # Check if the image is of sufficent quality for recognition.
    sufficientQuality = True
    detected_faces = face_client.face.detect_with_url(url=image, detection_model='detection_03', recognition_model='recognition_04', return_face_attributes=['qualityForRecognition'])
    for face in detected_faces:
        if face.face_attributes.quality_for_recognition != QualityForRecognition.high:
            sufficientQuality = False
            break
        face_client.person_group_person.add_face_from_url(PERSON_GROUP_ID, man.person_id, image)
        print("face {} added to person {}".format(face.face_id, man.person_id))

    if not sufficientQuality: continue

# Add to child person
for image in child_images:
    # Check if the image is of sufficent quality for recognition.
    sufficientQuality = True
    detected_faces = face_client.face.detect_with_url(url=image, detection_model='detection_03', recognition_model='recognition_04', return_face_attributes=['qualityForRecognition'])
    for face in detected_faces:
        if face.face_attributes.quality_for_recognition != QualityForRecognition.high:
            sufficientQuality = False
            print("{} has insufficient quality".format(face))
            break
        face_client.person_group_person.add_face_from_url(PERSON_GROUP_ID, child.person_id, image)
        print("face {} added to person {}".format(face.face_id, child.person_id))
    if not sufficientQuality: continue


'''
Train PersonGroup
'''
# Train the person group
print("pg resource is {}".format(PERSON_GROUP_ID))
rawresponse = face_client.person_group.train(PERSON_GROUP_ID, raw= True)
print(rawresponse)

while (True):
    training_status = face_client.person_group.get_training_status(PERSON_GROUP_ID)
    print("Training status: {}.".format(training_status.status))
    print()
    if (training_status.status is TrainingStatusType.succeeded):
        break
    elif (training_status.status is TrainingStatusType.failed):
        face_client.person_group.delete(person_group_id=PERSON_GROUP_ID)
        sys.exit('Training the person group has failed.')
    time.sleep(5)

'''
Identify a face against a defined PersonGroup
'''
# Group image for testing against
test_image = "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/identification1.jpg"

print('Pausing for 10 seconds to avoid triggering rate limit on free account...')
time.sleep (10)

# Detect faces
face_ids = []
# We use detection model 3 to get better performance, recognition model 4 to support quality for recognition attribute.
faces = face_client.face.detect_with_url(test_image, detection_model='detection_03', recognition_model='recognition_04', return_face_attributes=['qualityForRecognition'])
for face in faces:
    # Only take the face if it is of sufficient quality.
    if face.face_attributes.quality_for_recognition == QualityForRecognition.high or face.face_attributes.quality_for_recognition == QualityForRecognition.medium:
        face_ids.append(face.face_id)

# Identify faces
results = face_client.face.identify(face_ids, PERSON_GROUP_ID)
print('Identifying faces in image')
if not results:
    print('No person identified in the person group')
for identifiedFace in results:
    if len(identifiedFace.candidates) > 0:
        print('Person is identified for face ID {} in image, with a confidence of {}.'.format(identifiedFace.face_id, identifiedFace.candidates[0].confidence)) # Get topmost confidence score

        # Verify faces
        verify_result = face_client.face.verify_face_to_person(identifiedFace.face_id, identifiedFace.candidates[0].person_id, PERSON_GROUP_ID)
        print('verification result: {}. confidence: {}'.format(verify_result.is_identical, verify_result.confidence))
    else:
        print('No person identified for face ID {} in image.'.format(identifiedFace.face_id))
 

print()
print('End of quickstart.')

  3. Run your face recognition app from the application directory with the python command.

    python quickstart-file.py

    Tip

    The Face API runs on a set of pre-built models that are static by nature (the model's performance will not regress or improve as the service is run). The results that the model produces might change if Microsoft updates the model's backend without migrating to an entirely new model version. To take advantage of a newer version of a model, you can retrain your PersonGroup, specifying the newer model as a parameter with the same enrollment images.

Output

Person group: c8e679eb-0b71-43b4-aa91-ab8200cae7df
face 861d769b-d014-40e8-8b4a-7fd3bc9b425b added to person f80c1cfa-b8cb-46f8-9f7f-e72fbe402bc3
face e3c356a4-1ac3-4c97-9219-14648997f195 added to person f80c1cfa-b8cb-46f8-9f7f-e72fbe402bc3
face f9119820-c374-4c4d-b795-96ae2fec5069 added to person be4084a7-0c7b-4cf9-9463-3756d2e28e17
face 67d626df-3f75-4801-9364-601b63c8296a added to person be4084a7-0c7b-4cf9-9463-3756d2e28e17
face 19e2e8cc-5029-4087-bca0-9f94588fb850 added to person 3ff07c65-6193-4d3e-bf18-d7c106393cd5
face dcc61e80-16b1-4241-ae3f-9721597bae4c added to person 3ff07c65-6193-4d3e-bf18-d7c106393cd5
pg resource is c8e679eb-0b71-43b4-aa91-ab8200cae7df
<msrest.pipeline.ClientRawResponse object at 0x00000240DAD47310>
Training status: running.

Training status: succeeded.

Pausing for 10 seconds to avoid triggering rate limit on free account...
Identifying faces in image
Person for face ID 40582995-d3a8-41c4-a9d1-d17ae6b46c5c is identified in image, with a confidence of 0.96725.
Person for face ID 7a0368a2-332c-4e7a-81c4-2db3d74c78c5 is identified in image, with a confidence of 0.96921.
No person identified for face ID c4a3dd28-ef2d-457e-81d1-a447344242c4 in image.
Person for face ID 360edf1a-1e8f-402d-aa96-1734d0c21c1c is identified in image, with a confidence of 0.92886.

Clean up resources

If you want to clean up and remove an Azure AI services subscription, you can delete the resource or resource group. Deleting the resource group also deletes any other resources associated with it.

To delete the PersonGroup you created in this quickstart, run the following code in your script:

# Delete the main person group.
face_client.person_group.delete(person_group_id=PERSON_GROUP_ID)
print("Deleted the person group {} from the source location.".format(PERSON_GROUP_ID))
print()

Next steps

In this quickstart, you learned how to use the Face client library for Python to do basic face identification. Next, learn about the different face detection models and how to specify the right model for your use case.

Specify a face detection model version

Get started with facial recognition using the Face client library for JavaScript. Follow these steps to install the package and try out the example code for basic tasks. The Face service provides you with access to advanced algorithms for detecting and recognizing human faces in images. Follow these steps to install the package and try out the example code for basic face identification using remote images.

Reference documentation | Library source code | Package (npm) | Samples

Prerequisites

  • Azure subscription - Create one for free
  • The latest version of Node.js
  • Your Azure account must have a Cognitive Services Contributor role assigned in order for you to agree to the responsible AI terms and create a resource. To get this role assigned to your account, follow the steps in the Assign roles documentation, or contact your administrator.
  • Once you have your Azure subscription, Create a Face resource in the Azure portal to get your key and endpoint. After it deploys, select Go to resource.
    • You'll need the key and endpoint from the resource you create to connect your application to the Face API.
    • You can use the free pricing tier (F0) to try the service, and upgrade later to a paid tier for production.

Create environment variables

In this example, write your credentials to environment variables on the local machine that runs the application.

Go to the Azure portal. If the resource you created in the Prerequisites section deployed successfully, select Go to resource under Next Steps. You can find your key and endpoint under Resource Management in the Keys and Endpoint page. Your resource key isn't the same as your Azure subscription ID.

Tip

Don't include the key directly in your code, and never post it publicly. See the Azure AI services security article for more authentication options like Azure Key Vault.

To set the environment variable for your key and endpoint, open a console window and follow the instructions for your operating system and development environment.

  1. To set the VISION_KEY environment variable, replace your-key with one of the keys for your resource.
  2. To set the VISION_ENDPOINT environment variable, replace your-endpoint with the endpoint for your resource.
setx VISION_KEY your-key

setx VISION_ENDPOINT your-endpoint

After you add the environment variables, you may need to restart any running programs that will read the environment variables, including the console window.

Identify and verify faces

  1. Create a new Node.js application

    In a console window (such as cmd, PowerShell, or Bash), create a new directory for your app, and navigate to it.

    mkdir myapp && cd myapp

    Run the npm init command to create a node application with a package.json file.

    npm init

  2. Install the @azure-rest/ai-vision-face npm packages:

    npm install @azure-rest/ai-vision-face

    Your app's package.json file is updated with the dependencies.

  3. Create a file named index.js, open it in a text editor, and paste in the following code:

    Note

    If you haven't received access to the Face service using the intake form, some of these functions won't work.

    const { randomUUID } = require("crypto");

const { AzureKeyCredential } = require("@azure/core-auth");

const createFaceClient = require("@azure-rest/ai-vision-face").default,
  { FaceAttributeTypeRecognition04, getLongRunningPoller } = require("@azure-rest/ai-vision-face");

/**
 * NOTE This sample might not work with the free tier of the Face service because it might exceed the rate limits.
 * If that happens, try inserting calls to sleep() between calls to the Face service.
 */
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

const main = async () => {
  const endpoint = process.env["FACE_ENDPOINT"] ?? "<endpoint>";
  const apikey = process.env["FACE_APIKEY"] ?? "<apikey>";
  const credential = new AzureKeyCredential(apikey);
  const client = createFaceClient(endpoint, credential);

  const imageBaseUrl =
    "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/";
  const personGroupId = randomUUID();

  console.log("========IDENTIFY FACES========");
  console.log();

  // Create a dictionary for all your images, grouping similar ones under the same key.
  const personDictionary = {
    "Family1-Dad": ["Family1-Dad1.jpg", "Family1-Dad2.jpg"],
    "Family1-Mom": ["Family1-Mom1.jpg", "Family1-Mom2.jpg"],
    "Family1-Son": ["Family1-Son1.jpg", "Family1-Son2.jpg"],
    "Family1-Daughter": ["Family1-Daughter1.jpg", "Family1-Daughter2.jpg"],
    "Family2-Lady": ["Family2-Lady1.jpg", "Family2-Lady2.jpg"],
    "Family2-Man": ["Family2-Man1.jpg", "Family2-Man2.jpg"],
  };

  // A group photo that includes some of the persons you seek to identify from your dictionary.
  const sourceImageFileName = "identification1.jpg";

  // Create a person group.
  console.log(`Creating a person group with ID: ${personGroupId}`);
  await client.path("/persongroups/{personGroupId}", personGroupId).put({
    body: {
      name: personGroupId,
      recognitionModel: "recognition_04",
    },
  });

  // The similar faces will be grouped into a single person group person.
  console.log("Adding faces to person group...");
  await Promise.all(
    Object.keys(personDictionary).map(async (name) => {
      console.log(`Create a persongroup person: ${name}`);
      const createPersonGroupPersonResponse = await client
        .path("/persongroups/{personGroupId}/persons", personGroupId)
        .post({
          body: { name },
        });

      const { personId } = createPersonGroupPersonResponse.body;

      await Promise.all(
        personDictionary[name].map(async (similarImage) => {
          // Check if the image is of sufficent quality for recognition.
          const detectResponse = await client.path("/detect").post({
            contentType: "application/json",
            queryParameters: {
              detectionModel: "detection_03",
              recognitionModel: "recognition_04",
              returnFaceId: false,
              returnFaceAttributes: [FaceAttributeTypeRecognition04.QUALITY_FOR_RECOGNITION],
            },
            body: { url: `${imageBaseUrl}${similarImage}` },
          });

          const sufficientQuality = detectResponse.body.every(
            (face) => face.faceAttributes?.qualityForRecognition === "high",
          );
          if (!sufficientQuality) {
            return;
          }

          // Quality is sufficent, add to group.
          console.log(
            `Add face to the person group person: (${name}) from image: (${similarImage})`,
          );
          await client
            .path(
              "/persongroups/{personGroupId}/persons/{personId}/persistedfaces",
              personGroupId,
              personId,
            )
            .post({
              queryParameters: { detectionModel: "detection_03" },
              body: { url: `${imageBaseUrl}${similarImage}` },
            });
        }),
      );
    }),
  );
  console.log("Done adding faces to person group.");

  // Start to train the person group.
  console.log();
  console.log(`Training person group: ${personGroupId}`);
  const trainResponse = await client
    .path("/persongroups/{personGroupId}/train", personGroupId)
    .post();
  const poller = await getLongRunningPoller(client, trainResponse);
  await poller.pollUntilDone();
  console.log(`Training status: ${poller.getOperationState().status}`);
  if (poller.getOperationState().status !== "succeeded") {
    return;
  }

  // Detect faces from source image url and only take those with sufficient quality for recognition.
  const detectResponse = await client.path("/detect").post({
    contentType: "application/json",
    queryParameters: {
      detectionModel: "detection_03",
      recognitionModel: "recognition_04",
      returnFaceId: true,
    },
    body: { url: `${imageBaseUrl}${sourceImageFileName}` },
  });
  const faceIds = detectResponse.body.map((face) => face.faceId);

  // Identify the faces in a person group.
  const identifyResponse = await client.path("/identify").post({
    body: { faceIds, personGroupId },
  });
  await Promise.all(
    identifyResponse.body.map(async (result) => {
      try {
        const getPersonGroupPersonResponse = await client
          .path(
            "/persongroups/{personGroupId}/persons/{personId}",
            personGroupId,
            result.candidates[0].personId,
          )
          .get();
        const person = getPersonGroupPersonResponse.body;
        console.log(
          `Person: ${person.name} is identified for face in: ${sourceImageFileName} with ID: ${result.faceId}. Confidence: ${result.candidates[0].confidence}`,
        );

        // Verification:
        const verifyResponse = await client.path("/verify").post({
          body: {
            faceId: result.faceId,
            personGroupId,
            personId: person.personId,
          },
        });
        console.log(
          `Verification result between face ${result.faceId} and person ${person.personId}: ${verifyResponse.body.isIdentical} with confidence: ${verifyResponse.body.confidence}`,
        );
      } catch (error) {
        console.log(`No persons identified for face with ID ${result.faceId}`);
      }
    }),
  );
  console.log();

  // Delete person group.
  console.log(`Deleting person group: ${personGroupId}`);
  await client.path("/persongroups/{personGroupId}", personGroupId).delete();
  console.log();

  console.log("Done.");
};

main().catch(console.error);

  4. Run the application with the node command on your quickstart file.

    node index.js

Output

========IDENTIFY FACES========

Creating a person group with ID: c08484e0-044b-4610-8b7e-c957584e5d2d
Adding faces to person group...
Create a persongroup person: Family1-Dad.
Create a persongroup person: Family1-Mom.
Create a persongroup person: Family2-Lady.
Create a persongroup person: Family1-Son.
Create a persongroup person: Family1-Daughter.
Create a persongroup person: Family2-Man.
Add face to the person group person: (Family1-Son) from image: Family1-Son2.jpg.
Add face to the person group person: (Family1-Dad) from image: Family1-Dad2.jpg.
Add face to the person group person: (Family1-Mom) from image: Family1-Mom1.jpg.
Add face to the person group person: (Family2-Man) from image: Family2-Man1.jpg.
Add face to the person group person: (Family1-Son) from image: Family1-Son1.jpg.
Add face to the person group person: (Family2-Lady) from image: Family2-Lady2.jpg.
Add face to the person group person: (Family1-Mom) from image: Family1-Mom2.jpg.
Add face to the person group person: (Family1-Dad) from image: Family1-Dad1.jpg.
Add face to the person group person: (Family2-Man) from image: Family2-Man2.jpg.
Add face to the person group person: (Family2-Lady) from image: Family2-Lady1.jpg.
Done adding faces to person group.

Training person group: c08484e0-044b-4610-8b7e-c957584e5d2d.
Training status: succeeded.
No persons identified for face with ID 259dd648-be70-499c-9942-3512594e21eb
Person: Family1-Mom is identified for face in: identification1.jpg with ID: b7f7f542-c338-4a40-ad52-e61772bc6e14. Confidence: 0.96921.
Person: Family1-Son is identified for face in: identification1.jpg with ID: 600dc1b4-b2c4-4516-87de-edbbdd8d7632. Confidence: 0.92886.
Person: Family1-Dad is identified for face in: identification1.jpg with ID: e83b494f-9ad2-473f-9d86-3de79c01e345. Confidence: 0.96725.
Verification result between face bb7f7f542-c338-4a40-ad52-e61772bc6e14 and person de1d7dea-a393-4f69-9062-10cb66d4cf17: true with confidence: 0.96921
Verification result between face 600dc1b4-b2c4-4516-87de-edbbdd8d7632 and person 05fd84e4-41b0-4716-b767-4376e33fa207: true with confidence: 0.92886
Verification result between face e83b494f-9ad2-473f-9d86-3de79c01e345 and person c5124fe2-39dd-47ba-9163-1ed2998fdeb2: true with confidence: 0.96725

Deleting person group: c08484e0-044b-4610-8b7e-c957584e5d2d

Done.

Clean up resources

If you want to clean up and remove an Azure AI services subscription, you can delete the resource or resource group. Deleting the resource group also deletes any other resources associated with it.

Next steps

In this quickstart, you learned how to use the Face client library for JavaScript to do basic face identification. Next, learn about the different face detection models and how to specify the right model for your use case.

Specify a face detection model version

Get started with facial recognition using the Face REST API. The Face service provides you with access to advanced algorithms for detecting and recognizing human faces in images.

Note

This quickstart uses cURL commands to call the REST API. You can also call the REST API using a programming language. Complex scenarios like face identification are easier to implement using a language SDK. See the GitHub samples for examples in C#, Python, Java, JavaScript, and Go.

Prerequisites

  • Azure subscription - Create one for free
  • Your Azure account must have a Cognitive Services Contributor role assigned in order for you to agree to the responsible AI terms and create a resource. To get this role assigned to your account, follow the steps in the Assign roles documentation, or contact your administrator.
  • Once you have your Azure subscription, create a Face resource in the Azure portal to get your key and endpoint. After it deploys, select Go to resource.
    • You'll need the key and endpoint from the resource you create to connect your application to the Face API. You'll paste your key and endpoint into the code below later in the quickstart.
    • You can use the free pricing tier (F0) to try the service, and upgrade later to a paid tier for production.
  • PowerShell version 6.0+, or a similar command-line application.
  • cURL installed.

Identify and verify faces

Note

If you haven't received access to the Face service using the intake form, some of these functions won't work.

  1. First, call the Detect API on the source face. This is the face that we'll try to identify from the larger group. Copy the following command to a text editor, insert your own key and endpoint, and then copy it into a shell window and run it.

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/detect?returnFaceId=true&returnFaceLandmarks=false&recognitionModel=recognition_04&returnRecognitionModel=false&detectionModel=detection_03&faceIdTimeToLive=86400" -H "Content-Type: application/json" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data-ascii "{""url"":""https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/identification1.jpg""}"

    Save the returned face ID string to a temporary location. You'll use it again at the end.

  2. Next you'll need to create a LargePersonGroup and give it an arbitrary ID that matches regex pattern ^[a-z0-9-_]+$. This object will store the aggregated face data of several persons. Run the following command, inserting your own key. Optionally, change the group's name and metadata in the request body.

    curl.exe -v -X PUT "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}" -H "Content-Type: application/json" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data-ascii "{
    ""name"": ""large-person-group-name"",
    ""userData"": ""User-provided data attached to the large person group."",
    ""recognitionModel"": ""recognition_04""
}"

    Save the specified ID of the created group to a temporary location.

  3. Next, you'll create Person objects that belong to the group. Run the following command, inserting your own key and the ID of the LargePersonGroup from the previous step. This command creates a Person named "Family1-Dad".

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}/persons" -H "Content-Type: application/json" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data-ascii "{
    ""name"": ""Family1-Dad"",
    ""userData"": ""User-provided data attached to the person.""
}"

    After you run this command, run it again with different input data to create more Person objects: "Family1-Mom", "Family1-Son", "Family1-Daughter", "Family2-Lady", and "Family2-Man".

    Save the IDs of each Person created; it's important to keep track of which person name has which ID.

  4. Next you'll need to detect new faces and associate them with the Person objects that exist. The following command detects a face from the image Family1-Dad1.jpg and adds it to the corresponding person. You need to specify the personId as the ID that was returned when you created the "Family1-Dad" Person object. The image name corresponds to the name of the created Person. Also enter the LargePersonGroup ID and your key in the appropriate fields.

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}/persons/{personId}/persistedfaces?detectionModel=detection_03" -H "Content-Type: application/json" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data-ascii "{""url"":""https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/Face/images/Family1-Dad1.jpg""}"

    Then, run the above command again with a different source image and target Person. The images available are: Family1-Dad1.jpg, Family1-Dad2.jpg Family1-Mom1.jpg, Family1-Mom2.jpg, Family1-Son1.jpg, Family1-Son2.jpg, Family1-Daughter1.jpg, Family1-Daughter2.jpg, Family2-Lady1.jpg, Family2-Lady2.jpg, Family2-Man1.jpg, and Family2-Man2.jpg. Be sure that the Person whose ID you specify in the API call matches the name of the image file in the request body.

    At the end of this step, you should have multiple Person objects that each have one or more corresponding faces, detected directly from the provided images.

  5. Next, train the LargePersonGroup with the current face data. The training operation teaches the model how to associate facial features, sometimes aggregated from multiple source images, to each single person. Insert the LargePersonGroup ID and your key before running the command.

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}/train" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data ""

  6. Check whether the training status is succeeded. If not, wait for a while and query again.

    curl.exe -v "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}/training" -H "Ocp-Apim-Subscription-Key: {subscription key}"

  7. Now you're ready to call the Identify API, using the source face ID from the first step and the LargePersonGroup ID. Insert these values into the appropriate fields in the request body, and insert your key.

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/identify" -H "Content-Type: application/json" -H "Ocp-Apim-Subscription-Key: {subscription key}" --data-ascii "{
    ""largePersonGroupId"": ""INSERT_PERSONGROUP_ID"",
    ""faceIds"": [
        ""INSERT_SOURCE_FACE_ID""
    ],
    ""maxNumOfCandidatesReturned"": 1,
    ""confidenceThreshold"": 0.5
}"

    The response should give you a Person ID indicating the person identified with the source face. It should be the ID that corresponds to the "Family1-Dad" person, because the source face is of that person.

  8. To do face verification, you'll use the Person ID returned in the previous step, the LargePersonGroup ID, and also the source face ID. Insert these values into the fields in the request body, and insert your key.

    curl.exe -v -X POST "https://{resource endpoint}/face/v1.0/verify" `
-H "Content-Type: application/json" `
-H "Ocp-Apim-Subscription-Key: {subscription key}" `
--data-ascii "{
    ""faceId"": ""INSERT_SOURCE_FACE_ID"",
    ""personId"": ""INSERT_PERSON_ID"",
    ""largePersonGroupId"": ""INSERT_PERSONGROUP_ID""
}"

    The response should give you a boolean verification result along with a confidence value.

Clean up resources

To delete the LargePersonGroup you created in this exercise, run the LargePersonGroup - Delete call.

curl.exe -v -X DELETE "https://{resource endpoint}/face/v1.0/largepersongroups/{largePersonGroupId}" -H "Ocp-Apim-Subscription-Key: {subscription key}"

If you want to clean up and remove an Azure AI services subscription, you can delete the resource or resource group. Deleting the resource group also deletes any other resources associated with it.

Next steps

In this quickstart, you learned how to use the Face REST API to do basic facial recognition tasks. Next, learn about the different face detection models and how to specify the right model for your use case.

Specify a face detection model version