What is Azure Face in Foundry Tools?
Azure Face is a Foundry Tools service that provides AI algorithms to detect, recognize, and analyze human faces in images. According to Microsoft, typical scenarios include identification, touchless access control, and automatic face blurring for privacy.
Access to the service is deliberately restricted. Microsoft makes Azure Face available only to Microsoft managed customers and partners; access is reviewed through an intake form. The client SDKs for liveness detection are a separately gated feature as well. These restrictions are part of Microsoft’s Responsible AI requirements.
The service can be used through client library SDKs or by calling the REST API directly. Face detection is always the first step: the Detect API returns the rectangle coordinates of detected faces plus a unique ID representing the stored face data, which is then used in downstream operations such as verification or identification.
Core Features
- Face detection with rectangle coordinates and a unique face ID, optionally with attributes such as head pose
- Verification (one-to-one): confirming whether two faces belong to the same person
- Identification (one-to-many): matching a face against a secure repository of people
- Liveness detection as an anti-spoofing measure against printouts, masks, and screen replays
- Find Similar with matchPerson and matchFace modes for face search by image
- Group operation to divide unknown faces into smaller groups based on similarity
- Client library SDKs and REST API; liveness SDKs for Android, iOS, and Web
Typical Use Cases
Identity verification
A person is verified against a trusted reference image, for instance from a government-issued ID or an enrollment photo taken in person. Liveness detection confirms the image comes from a physically present person rather than a printout or mask.
Touchless access control
Compared with cards or tickets, opt-in face identification enables a more convenient access control experience and reduces hygiene and security risks from card sharing, loss, or theft. Microsoft names check-in scenarios with a human in the loop, such as airports, stadiums, buildings, reception kiosks, hospitals, or schools.
Privacy protection in video material
Detected faces in recorded video can be redacted or blurred to protect the privacy of the people recorded.
Benefits
- Verification and identification through a single API, combinable with liveness detection
- Demonstrated robustness of liveness detection: 0% penetration rate in iBeta Level 1 and Level 2 PAD tests conformant to ISO/IEC 30107-3
- Scalable repositories with up to 1 million person objects per group
- SDKs for Android, iOS, and Web for integration into mobile and web applications
- Responsible AI guardrails through limited access and retired or restricted attributes
Integration with innFactory
As a Microsoft Solutions Partner, innFactory supports projects involving Azure Face: we accompany the limited access application, assess the data protection requirements for processing biometric data, and design the architecture and integration for verification or access control scenarios.
We also examine with you whether a project can avoid biometric processing altogether, and support the implementation of consent, retention, and deletion concepts. Contact us for a no-obligation consultation.
Typical Use Cases
Technical Specifications
Frequently Asked Questions
Can any Azure customer use Azure Face?
No. Microsoft states: '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.' Access is requested through the Face Recognition intake form. The liveness client SDKs are also a gated feature requiring a separate request.
What does liveness detection do?
Liveness detection determines whether a face in an input video stream is real or fake and prevents spoofing via photos, videos, or masks. Microsoft states the solution achieved a 0% penetration rate in iBeta Level 1 and Level 2 Presentation Attack Detection tests conducted by a NIST/NVLAP-accredited laboratory and conformant to ISO/IEC 30107-3.
Which face attributes are still available?
Microsoft has retired or limited capabilities that could be used to infer emotional states and identity attributes. Emotion and gender were retired. Age, smile, facial hair, hair, and makeup are limited; using them requires contacting the Azure Face team.
What obligations apply when processing biometric data?
Microsoft notes that when you use Microsoft products or services to process biometric data, you are responsible for providing notice to data subjects including retention periods and destruction, obtaining consent, and deleting the biometric data, all as appropriate and required under applicable data protection requirements.
Are there restrictions on use by government agencies?
Yes. On June 11, 2020, Microsoft announced it will not sell facial recognition technology to police departments in the United States until strong regulation grounded in human rights has been enacted. When creating a new Face resource, customers must acknowledge in the Azure portal that they will not use the service by or for a police department in the United States.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Azure (official documentation). This page does not represent an offer by Azure.
