All Products
Search
Document Center

ID Verification:FACE_IDU_MIN

Last Updated:Jul 13, 2026

FACE_IDU_MIN is an API-based real-person face detection service. The service accepts pre-acquired face images as input and uses the Qwen-VL large model to deeply analyze forgery risks and accurately determine whether the subject is a real person. The service supports multiple comparison and verification solutions: 1:1 identity verification against a stored face image, and 1:N search in a face group to determine whether a person exists. After successful verification, the service can also automatically register the face to a specified face group

API description

  • API operation: FaceVerifyIntl

  • Request method: POST over HTTPS

  • Description: Call the FaceVerifyIntl API to perform liveness verification.

  • QPS limit: Each API has a dedicated QPS limit. For more information, see QPS limits of ID Verification server-side APIs.

  • Endpoints:

    Note

    An internal network refers to the internal communication network between Alibaba Cloud services in the same region. If your business server is deployed in the corresponding Alibaba Cloud region, you can use the internal endpoint to access the ID Verification service for more secure and stable network communication.

    Singapore

    • Public endpoint: cloudauth-intl.ap-southeast-1.aliyuncs.com

    • Internal endpoint: cloudauth-intl-vpc.ap-southeast-1.aliyuncs.com

    Indonesia (Jakarta)

    • Public endpoint: cloudauth-intl.ap-southeast-5.aliyuncs.com

    • Internal endpoint: cloudauth-intl-vpc.ap-southeast-5.aliyuncs.com

    China (Hong Kong)

    • Public endpoint: cloudauth-intl.cn-hongkong.aliyuncs.com

    • Internal endpoint: cloudauth-intl-vpc.cn-hongkong.aliyuncs.com

    Malaysia (Kuala Lumpur)

    • Public endpoint: cloudauth-intl.ap-southeast-3.aliyuncs.com

    • Internal endpoint: cloudauth-intl-vpc.ap-southeast-3.aliyuncs.com

Online debugging and integration

Note

Before debugging and integrating, make sure that you have thoroughly read the Use OpenAPI documentation to fully understand how to call the API on the OpenAPI platform and how to obtain SDKs and code samples.

You can directly run this API in OpenAPI Explorer for debugging, and generate SDK code samples for this API.

Request parameters

Face image parameters

  • Three methods are available to provide a face image. Choose any one of them.

    • Base64 mode: SourceFacePicture/TargetFacePicture

    • URL mode: SourceFacePictureUrl/TargetFacePictureUrl

    • File stream mode: SourceFacePictureFile/TargetFacePictureFile

  • Image format: JPG, JPEG, or PNG.

  • Image size: Recommended 50–100 KB, maximum 1 MB.

  • Image resolution: Maximum 1920*1080 (height × width), minimum 640*480 (height × width). Scale the shorter side to 720 px with compression ratio > 0.9. Height must exceed width; landscape orientation may reduce accuracy.

    Note

    Base64 encoding inflates file size. Keep the original image under 0.6 MB to stay within the 1 MB transfer limit.

  • Image quality recommendations:

    • Face must be complete, clear, unobstructed, and front-facing. Use a front-facing camera.

    • Face must occupy more than 60% of the image area. Smaller faces reduce accuracy.

    • Multiple faces: the algorithm processes the largest one. Avoid multi-face images.

Request parameters

Name

Type

Required

Description

Example

ProductCode

String

Yes

The product code. Set the value to FACE_IDU_MIN.

FACE_IDU_MIN

MerchantBizId

String

Yes

A custom business-unique identifier used for subsequent issue locating and troubleshooting. The value supports a combination of letters and digits with a length of 32 characters. Make sure that the value is unique.

e0c34a77f5ac40a5aa5e6ed20c35****

MerchantUserId

String

Yes

A custom user ID, or another identifier that can identify a specific user, such as a mobile number or email address. We strongly recommend that you desensitize the value in advance, for example, by hashing it.

123456789

VerifyModel

String

Yes

The verification mode:

  • 0: Search mode

    • Function: Pass in a face group and a user face image (SourceFacePicture/SourceFacePictureUrl/FacePictureFile). The system automatically searches the face group to determine whether the face image already exists. The user face image supports silent liveness detection.

    • Recommended scenario: Registering real-person accounts where duplicate registration is not allowed.

  • 1: Verification mode (default)

    • Function: Pass in a specified face image (SourceFacePicture/SourceFacePictureUrl/FacePictureFile) and a reference face image (TargetFacePicture/TargetFacePictureUrl/TargetFacePictureFile). The system automatically verifies whether the face information of the two images matches. The specified face image supports silent liveness detection.

    • Recommended scenario: Scenarios where you need to verify the operator's identity when modifying login credentials, account information, etc.

  • 2: Comprehensive mode

    • Function: Pass in a face group, a specified face image (SourceFacePicture/SourceFacePictureUrl/FacePictureFile), and a reference face image (TargetFacePicture/TargetFacePictureUrl/TargetFacePictureFile). The system automatically searches the face group for the specified face image, checks whether it matches the reference face image, and supports silent liveness detection for the specified face image.

    • Recommended scenario: Scenarios where you need to verify that the user is both new and the legitimate operator.

0

FaceGroupCodes

String

No

Obtain the corresponding code when you create a face group in the ID Verification console. You can query up to 10 face groups at the same time. Separate multiple face group codes with commas (,).

1232344,23444

SourceFacePicture

String

No

The Base64-encoded face image.

base64

SourceFacePictureUrl

String

No

The URL of the face image, which must be publicly accessible over HTTP or HTTPS.

https://***face1.jpeg

SourceFacePictureFile

String

No

The face image file stream.

Face image file stream 1

TargetFacePicture

String

No

The Base64-encoded reference (target) face image.

base64

TargetFacePictureUrl

String

No

The URL of the reference (target) face image, which must be publicly accessible over HTTP or HTTPS.

https://***face2.jpeg

TargetFacePictureFile

String

No

The reference (target) face image file stream.

Face image file stream 2

AutoRegistration

String

No

Specifies whether to automatically register the face to a specified face group when the face is not found during search.

  • 0: Automatic registration

  • 1: Do not register (default)

0

FaceRegisterGroupCode

String

No

The face group for registration.

0e0c34a77f

ReturnFaces

String

No

The number of faces to return when multiple faces exist above the matching threshold.

  • Default value: 1

  • Maximum value: 5

1

FaceQualityCheck

String

No

Enables face quality inspection. Disabled by default. Valid values:

  • N: Disabled (default).

  • Y: Enabled.

When enabled, the system blocks low-quality photos (blurry, occluded) from comparison. Recommended when higher accuracy is required or to reduce invalid requests. Returns 401 (UnqualifiedPhoto) for blocked images.

N

FaceAttributeCheck

String

No

Whether to return additional face attribute fields.

  • N: Disabled (default).

  • Y: Enabled. When enabled, the attributes are returned as a string in the faceAttributeInfo field within ExtFaceInfo.

N

Response data

Name

Type

Description

Example

HTTP Status Code

Integer

The HTTP status code.

200

HTTP Body

RequestId

String

The ID of the request.

130A2C10-B9EE-4D84-88E3-5384FF0****

Code

String

Return code: For more information, see Server-side HTTP status codes.

Success

Message

String

This section provides detailed descriptions of the return codes.

success

Result.FacePassed

String

The final authentication result. Valid values:

  • Y: Passed.

  • N: Failed.

Y

Result.FaceComparisonScore

Double

Returned when VerifyModel is set to 1 or 2. The comparison score of 1:1 verification. Valid values: 0 to 100.

98

Result.DuplicateFace

String

Returned when a duplicate face is found during search. The face ID, user ID, and comparison score of the corresponding face in the face group are returned.

[
  {
    "faceGroupCode": "sg7****uzt",
    "faceId": "f5a921*******9e792ec84c8f0ca592a",
    "merchantUserId":"face0005",
    "score":93.26
  }
]

Result.FaceRegistrationResult

Integer

The face registration result.

  • 0: Failed

  • 1: Successful

1

Result.FaceRegistrationId

String

Returned when automatic registration is enabled and the face is successfully registered. The corresponding FACEID is returned.

9e792******a592a

Result.FaceAttack

String

Indicates whether the captured face involves a liveness attack. Y indicates an attack, and N indicates no attack. Returned when silent liveness detection is enabled.

N

Result.FaceAttributeInfo

String

Detects the facial attributes of the main face in an input image. The main face is the one that is largest, clearest, and closest to the camera. The detection results are returned as a string. You can use this string to make decisions about the content.

The returned fields are defined as follows:

  • gender: The gender of the main entity. A value of 1 indicates male, and 0 indicates female.

  • glasses: Indicates whether the main entity is wearing glasses. A value of 1 indicates that glasses are detected, and 0 indicates that they are not.

  • hat: Indicates whether the main entity is wearing a hat. This includes headscarves or veils. A value of 1 indicates that a hat is detected, and 0 indicates that it is not.

  • mask: Indicates whether the main entity is wearing a mask. A value of 1 indicates that a mask is detected, and 0 indicates that it is not.

  • tattoo: Indicates whether the main entity has a visible tattoo. A value of 1 indicates that a tattoo is detected, and 0 indicates that it is not.

  • smoking: Indicates whether the main entity is smoking. A value of 1 indicates that smoking is detected, and 0 indicates that it is not.

  • others: Indicates whether people other than the main entity are present. A value of 1 indicates that other people are detected, and 0 indicates that they are not.

  • env: The environment of the main entity, such as an office, factory, hospital, or vehicle. The system currently supports returning office, hospital, factory, vehicle, or other.

  • expression: Indicates whether the main entity has an exaggerated expression. An exaggerated expression is a strong facial expression that is distinct from a natural or calm state. A value of 1 indicates that an exaggerated expression is detected, and 0 indicates that it is not.

  • age: The age of the main entity in years.

{
        gender: 1
        glasses: 0
        hat: 1
        mask: 1
        tattoo: 0
        smoking: 0
        others: 1
        env:office
        expression:1
        age:16
}

Result.SubCode

String

A description of the authentication result. For more information, see the SubCode section.

200

Result.TransactionId

String

The unique identifier of the authentication request.

4ab0b***cbde97

Result.ExtFaceInfo

Object

The liveness detection result information. For the JSON format, see the example on the right. For more information, see the ExtFaceInfo section.

{
  "FaceQualityScore": 66.3,
  "OcclusionScore": 100,
  "SharpnessScore": 70.3,
  "KaOcclusionScore": 80,
  "IlluminationScore": 90.79
}

SubCode

Error code

Billable

Description

200

Yes

Authentication passed.

204

Yes

Face mismatch. The faces may not belong to the same person, or the liveness photo quality is low.

205

Yes

Liveness risk detected.

233

Yes

Similar face detected.

ExtFaceInfo

Name

Type

Description

Example

FaceQualityScore

Double

The liveness face quality score. Valid values: 0 to 100. A higher value indicates better quality.

88.62

OcclusionScore

Double

A quality assessment sub-dimension: the occlusion score. Valid values: 0 to 100. A higher value indicates better quality.

99.99

KaOcclusionScore

Double

A quality assessment sub-dimension: the key area occlusion score. Valid values: 0 to 100. A higher value indicates better quality.

100

IlluminationScore

Double

A quality assessment sub-dimension: the illumination score. Valid values: 0 to 100. A higher value indicates better quality.

97.43

SharpnessScore

Double

A quality assessment sub-dimension: the image sharpness score. Valid values: 0 to 100. A higher value indicates better quality.

60.78