World leading technology

Presentation attack detection (PAD) and face matching are the two outstanding capabilities of iProov's core one-time biometric technology. We deliver these with exceptional usability through our patented user interface and Flashmarks. Our systems are resilient and adaptive. All our technology is our own IPR.

In this section you can read more about:

Sustainable adaptive security

Presentation attack detection (PAD) including liveness, forgery and replay defence and deep learning.

Simple user experience

The verification process and details on accessibility and inclusiveness.

Simple integration

App (Android/iOS SDKs), Browser (HTML5), REST API, UI customisation and ID Matcher orchestration.

Our patents

10 granted patents cover our user interface as well as our Flashmarks; our core presentation attack detection (PAD) technology.

Sustainable adaptive security

Our Flashmark technology

iProov has developed the world's most comprehensive and usable liveness detection technology - effective against highly sophisticated attacks including replicas and replays.

This is how it works:
  • A unique one-time colour sequence code is generated in iProov’s servers and sent to the user’s device.
  • As the authentication begins, the user’s screen flashes for 2.5 seconds with the unique colour sequence; the Flashmarks.
  • The Flashmarks illuminate and reflect off the user’s face whilst a video is streamed back to iProov’s servers.
  • In the servers, iProov’s DELFINA matching engine compares the user’s Flashmarked image against their enrolment image and images of every prior or subsequent successful authentication.
  • A sophisticated in-depth analysis of the Flashmarked image, such as the pattern and characteristics of reflection, is then carried out to determine liveness.
  • The Flashmark colour sequence is also checked to ensure it corresponds to the unique one-time code sent to the user’s device to initiate the session.
  • Other aspects of the video image and other data are also analysed.
  • Flashmark colour sequences are never repeated in a user’s lifetime, creating one-time biometrics.
  • Flashmarked images cannot be reused and are worthless if stolen.

Presentation attack detection (PAD)

Attacks can be either physical or digital. A physical attack seeks to subvert biometric scanning time, such as using a 3D-printed mask or stolen still or video imagery. A digital attack targets the systems running during the authentication process and seeks to trigger a ‘positive’ match.

Attackers do not stay still and neither do we. Every claim is analysed by our servers; every attack is observed. When attackers experiment, we learn. When they innovate, so do we.

Our adaptive defences continually monitor and react to the evolving landscape of threats, keeping our customers safe from unseen, large-scale, persistent threats.

Replica and replay defence

There is only ever one genuine user face. The test for genuineness guarantees the security of facial authentication. This is what presentation attack detection and defence is for.

Good defence requires reliable and sustainable detection of two kinds of attacks.

1. Replicas

Realistic copies of faces are increasingly easy to make. High resolution images can easily be animated with blinking and movement, and even with expressions and speaking to create convincing imagery on a screen.

Device motion and parallax can be replicated, and faces can be digitally draped on other faces. Replica heads can now be printed from 3D models reconstructed from a few images.

2. Replays

Real imagery of a user authenticating under normal conditions is very dangerous because it will pass every replica test; it can be obtained by an attacker by methods including social engineering and device compromise.

Deep learning

Face matching is the core of our authentication. Genuine users must enjoy a simple, reliable experience, whilst impersonations are prevented.

We developed our DELFINA face matcher, which uses the latest advances in deep neural networks, to deliver world-leading matching performance. It is optimised for ‘selfies’ taken on smartphones and PCs in a huge variety of lighting conditions, poses and facial features.

Simple user experience
Verifier

  • Service providers use their own method to establish trust in a user, then enrol them using Verifier.
  • Whenever the user wants to subsequently gain access, they authenticate with Verifier.
  • On their device, an API enables the local app to contact Verifier.
  • Verifier returns a result token, which is then sent back to the user’s service provider.
  • The user’s service provider calls iProov’s servers using a secure REST API. This is used to confirm the token, preventing token forgery.
  • The authetication process is started and the user is verified in 2.5 seconds.
  • The user’s service provider refers to users by an anonymised pseudonym, thereby safeguarding the user's real-world identity.

Simple integration

Front-end integration:
  • Android SDK (v4.1 and higher)
  • iOS SDK (v9+ Swift and Objective-C)
  • Browser – HTML5 WebRTC (Chrome/Firefox/Safari/Edge)
  • Windows – proto-type client available on request
  • Backend integration with iProov REST API
  • UI customisation – all front-end SDKs come with an array of options to allow full customisation of the user experience to match your branding. For more information please consult the relevant SDK documentation on our Portal

Highly accessible and inclusive

In addition to being highly secure and intuitive, using iProov requires no complex written instructions as it offers clear guidance on face positioning. This also makes it well adapted for multi-language deployment.

Also, unlike most other face biometrics, iProov's Presentation attack detection (PAD) operates on passive challenge/response protocols. This means that in order to authenticate, users are not required to move their head from side-to-side, blink, or read out numbers; iProov's Flashmark technology determines liveness without the need for movement or speech.

Simple user experience
ID Matcher

  • Service providers use their own method to establish trust in a user, then enrol them using Verifier.
  • Whenever the user wants to subsequently gain access, they authenticate with Verifier.
  • On their device, an API enables the local app to contact Verifier.
  • Verifier returns a result token, which is then sent back to the user’s service provider.
  • The user’s service provider calls iProov’s servers using a secure REST API. This is used to confirm the token, preventing token forgery.
  • The authetication process is started and the user is verified in 2.5 seconds.
  • The user’s service provider refers to users by an anonymised pseudonym, thereby safeguarding the user's real-world identity.

Simple integration

Front-end integration:
  • Android SDK (v4.1 and higher)
  • iOS SDK (v9+ Swift and Objective C)
  • Browser – HTML5 WebRTC (Chrome/Firefox/Safari/Edge)
  • Windows – proto-type client available on request
  • Backend integration with iProov REST API

  • UI customisation – all front-end SDKs come with an array of options to allow full customisation of the user experience to match your branding. For more information please consult the relevant SDK documentation on our Portal

  • Orchestration for NFC is provided via Innovalor S.A. and for optical via a selection of other partners
  • Easily integrates into host apps and with third party document capture technology

Highly accessible and inclusive

In addition to being highly secure and intuitive, using iProov requires no complex written instructions as it offers clear guidance on face positioning. This also makes it well adapted for multi-language deployment.

Also, unlike most other face biometrics, iProov's Presentation attack detection (PAD) operates on passive challenge/response protocols. This means that in order to authenticate, users are not required to move their head from side-to-side, blink, or read out numbers; iProov's Flashmark technology determines liveness without the need for movement or speech.

Our 10 patents

Online Pseudonym Verification and Identity Validation

  • GB 2544915
    18 October 2017
  • GB 2543673
    18 October 2017
  • GB 2543117
    14 June 2017
  • GB 2542449
    9 August 2017
  • GB 2501362
    22 March 2017
  • US 9773180
    26 September 2017
  • US 9621548
    11 April 2017
  • US 9479500
    25 October 2017
  • US 9075975
    7 July 2015

Real-time Visual Feedback for User Positioning with Respect to a Camera and a Display

  • US 9412169
    9 August 2016

Try Our Solution

Verifiers innovative technology can be tested for free!

iProov Limited

Elizabeth House
39 York Road
London
SE1 7NQ
United Kingdom

Contact Us

Tel: +44 20 7993 2379

Email: This email address is being protected from spambots. You need JavaScript enabled to view it.