Identity Verification using Amazon Rekognition

Verify user identity online using ML
Deploy a solution template
Contact us

Why Identity Verification?

In-person user identity verification is slow to scale, costly, and high friction for users. Machine learning (ML)–powered facial biometrics can help online user identity verification. Amazon Rekognition offers pretrained facial recognition and analysis capabilities that you can quickly add to your user onboarding and authentication workflows to verify opted-in users' identity online. No ML expertise is required. With Amazon Rekognition, you can onboard and authenticate users in seconds while detecting fraudulent or duplicate accounts. As a result, you can grow users faster, reduce fraud, and lower user verification costs.

Benefits of Identity Verification

Convert more visitors into customers by decreasing onboarding time and increasing user convenience. With Amazon Rekognition, you can verify users in seconds online anywhere in the world and scale from hundreds to millions of identity verifications per hour. Users can now access your services online without having to visit in person.

Reduce the time and cost of in-person identity verification by using Amazon Rekognition pretrained and customizable APIs. With Amazon Rekognition, you can onboard and authenticate users online without building and managing your own ML infrastructure.

Strengthen your fraud prevention capabilities by complementing password-based authentication with online visual identity verification. Guard against fraudulent account openings or transactions by comparing a user’s selfie picture with an identity document picture or your collection of existing users’ pictures.

 

Page Topics

General

General

Amazon Rekognition Face Liveness helps you verify that only real users, not bad actors using spoofs, can access your services. You can 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.

Amazon Rekognition Face Detection helps you detect that the user’s selfie picture is captured correctly. You can detect if a face is present in the picture. You can also use predicted attributes such as bounding box size, pose, brightness, sharpness, eyes open, mouth open, and eyeglasses worn to determine picture quality.

Amazon Rekognition Face Comparison helps you measure the similarity of two faces to help you determine if they are the same person. You can receive a similarity score prediction for a user’s selfie picture against their identity document picture in near real time.

Amazon Rekognition Face Index and Search helps you create a face collection of existing users and search new users' selfie pictures against all faces in your collection to detect duplicate or fraudulent account creation attempts.

Amazon Rekognition Object Detection helps you determine the type of user identity document such as driver’s license or passport. You can also use Amazon Rekognition Custom Labels to detect an identity document type unique to your region by training a custom ML model with a few annotated images.

Amazon Rekognition Text Detection helps you extract key pieces of text on an identification card, such as name, date of issue, age, and identification number. You can compare this information with the user application form data.

Customers

  • Aella Credit

    Aella Credit provides instant loans to individuals with a verifiable source of income in emerging markets using biometric, employer, and mobile phone data.

    Identity verification and validation have been a major challenge in emerging markets. The ability to properly identify users is a key hindrance in building credit for billions of people in emerging markets. Using Amazon Rekognition for identity verification on our mobile application has reduced verification errors significantly and given us the ability to scale. We can now detect and verify an individual’s identity in real time without any human intervention, thereby allowing faster access to our products. We tried various well-advertised solutions, but none of the popular alternatives could accurately map out various skin tones. Amazon Rekognition helped us effectively recognize faces of our customers in our markets. It also helped us with KYC in discovering overlapping profiles and duplicate datasets.

    Wale Akanbi, CTO & Co-Founder, Aella Credit

  • AU Small Finance Bank

    AU Small Finance Bank (AU Bank), India’s largest small finance bank (based on assets and liabilities), has been successfully onboarding new customers with video KYC since 2020, supporting now more than 2.7 million customers across 900 banking touchpoints.

    AWS provides the necessary scalability and reliability for the WorkApps platform, and our time to market and time to value have accelerated significantly with a cloud-based solution.

    Ankur Tripathi, Chief Information Officer, AU Small Finance Bank

  • Carbon

     

    Carbon is a digital financial service platform powered by OneFi that provides services to underbanked individuals in West Africa through an Android mobile app, which has over 900,000 downloads.

    In May 2016, Carbon launched its mobile app for its loan application process. With the mobile app, images are constantly being generated and consumed at faster rates than before. Carbon needed to meet its growing need for image analysis for fraud detection and risk analysis. We wanted to be able to identify if a human face was truly detected in an uploaded image and identify other labels such as gender and identity. We chose Amazon Rekognition because of its ease in adding image analysis to our mobile app and the accuracy of its facial analysis.

    Olawale Olaleye, Head of IT Infrastructure Engineering, OneFi

  • Software Colombia

    Software Colombia is a top-tier AI and ML software development company providing cutting-edge technology solutions globally, with a focus on innovation, quality, and client satisfaction on its more than 300 active projects.

    Read success story

     

    Our main challenge was to implement a strong, yet fast and accurate user authentication platform, and Amazon Rekognition and its Face Liveness detection API helped us achieve it. This new Amazon Rekognition API enabled us to build an in-house biometric facial recognition process to helps us mitigate identity spoofing attacks and risk by up to 95%, making our X509 digital certificates issuing and signature processes more secure and efficient. The ability to give our customers the option to authenticate and verify their identities by using a phone camera also makes our services more inclusive and available across regions.

    Alex Chacón, CEO, Software Colombia

  • Q5id

    Q5id provides consumers and businesses with a robust Proven Identity Management solution to help customers verify identities and secure organizations.

    Watch Q5ID testimonial video

    Q5id is focused on proving individual identities versus assuming their validity. Our goal is to deliver the highest level of assurance to identify and verify that individuals are who they say they are for our financial service clients and their customers. We’re achieving this by working with Amazon Rekognition Identity Verification APIs and its face recognition capabilities, then integrating our proprietary software to build our products and services. AWS helped us to improve and balance the facial recognition identification patterns we use to achieve a false acceptance rate of 1 in 933 billion – a number more than 100 times the world’s population.

    Becky Wanta, Chief Technology Officer, Q5id