Skip to main content

Amazon Rekognition

Amazon Rekognition Face Liveness

Detect real users and deter bad actors using spoofs in seconds during facial verification

Get started
Contact us

Why Amazon Rekognition Face Liveness?

Amazon Rekognition Face Liveness can detect whether real users, not bad actors using spoofs, can access your services. Amazon Rekognition Face Liveness analyzes a short selfie video to detect spoofs presented to the camera, such as printed photos, digital photos, digital videos, or 3D masks, as well as spoofs that bypass the camera, such as pre-recorded or deepfake videos. Face Liveness is a fully managed feature that can be easily added to your React web, native iOS, and native Android applications running on most devices with a front-facing camera. No infrastructure management, hardware-specific implementation, or machine learning expertise is required. The feature automatically scales up or down in response to demand, and you only pay for the face liveness checks you perform.

Use cases

Reduce fraudulent account creation on your service by validating new users with Face Liveness. For example, financial services customers can use Face Liveness and Amazon Rekognition face matching to verify user identity prior to opening an online account.

Strengthen the verification of high-value user activities, such as device change, password change, and money transfer with Face Liveness. For example, ride-sharing customers can use Face Liveness and face matching to verify driver identity before commencing a ride.

Deter underage users from accessing restricted content with Face Liveness. For example, online gaming or dating customers can use Face Liveness and age estimation from Amazon Rekognition Facial Analysis to verify user’s age before granting access.

Avoid bots from using your service with Face Liveness. For example, social media customers can use Face Liveness for posing real human checks to keep bots at bay.

Capabilities

Type
Benefits
Details
High accuracy
Presentation attacks detection

Bypass attacks detection

3D mask attacks detection

Configurable confidence score
Detects spoof attacks presented to the camera, such as printed 2D photos, 2D cut-out paper masks, and hi-res photos or videos on a digital screen.

Detects spoof attacks that bypass the camera, such as pre-recorded, synthetic, and deepfake videos directly injected into the video capture sub-system.

Detects spoof attacks that use 3D masks made of silicone, latex, plastic, cloth, and more.

Provides a confidence score between 0 and 100 for adjustable accuracy levels based on your use case.
Low user friction

Near passive user action

Fast verification

User guidance and feedback

Accessibility compliance

Requires simple action of moving face into an oval rendered on user's device screen similar to taking a selfie video.

Analyzes user selfie video in real-time to minimize end-to-end latency and deliver results in seconds.

Provides on-screen instructions and contextual guidance to help end-users complete liveness checks quickly.

Adheres to Web Content Accessibility Guidelines (WCAG) 2.1 for the colored screens displayed in the Face Liveness user challenge,
minimizing impact to photo-sensitive users.
Ease of integration

Broad platform support

Pre-built user interface (UI) components

Optimized video capture and streaming

Selfie frame with quality checks

Integrates into AWS Amplify SDKs for web (React) and mobile (native iOS and native Android). No hardware-specific implementation required.

Provides pre-built user interfaces to quickly integrate Face Liveness into your application.

Minimizes selfie video size to under 1 MB for efficient data transfer.

Provides a high-quality selfie frame for face matching or age estimation.
Ease of management

Fully managed

Audit Image

Pay per use and Autoscaling

Open-source device SDKs

No need to deploy or manage liveness sotware in their on-premises or hosted infrastructure.

Returns up to four frames for manual inspection or audit trail purposes.

Pay per liveness check and automatically scale up to millions of liveness checks per day.

Full transparency and visibility into AWS Amplify SDKs.

Software Colombia

AWS identity verification and its new Amazon Rekognition Face Liveness helped our new eLogic biometrical solution reduce fraud and risk by 95%, while making our product more inclusive and accessible.

Alex Chacón, CEO, Software Colombia
Missing alt text value

Page topics

General

Open all

Face Liveness is available in five AWS regions - US East (N. Virginia), US West (Oregon), Europe (Ireland), Asia Pacific (Tokyo), and Asia Pacific (Mumbai). To learn more about the region support, visit the Amazon Rekognition endpoints page.

The Face Liveness feature produces a probabilistic confidence score, ranging from 0 to 100. A higher score corresponds to a higher confidence that the user is live and real. Face Liveness provides a selfie frame for face matching or age estimation. The feature also returns up to four audit frames for human review and audit trail purposes.

We do not recommend using Face Liveness and face matching in place of username/password. We recommend using Face Liveness and face matching as a secondary or step-up method to username/password for an additional layer of security.

Face Liveness is trained and tested using datasets that represent a diverse range of human facial features and skin tones under a wide range of environmental variations. This includes datasets of selfie videos for which we have reliable demographic labels such as gender, age, and skin tone.

Face Liveness may store and use selfie video processed solely to provide, maintain, and improve the feature, unless you opt out. To learn more, visit Amazon Rekognition data privacy.

Face Liveness is priced on a per check basis. To learn more, visit the Amazon Rekognition pricing page.