SmartFace Gets Major Update for Improved Scalability


Innovatrics has recently released SmartFace 3.2, a major update which gives users even more flexibility for scaling. Additionally, this latest version makes it possible for every video stream to run in a separate process. With this added capability, increasing resource requirements is highly-manageable, particularly when adding new cameras into the system.

face recognition

Efficiently designed to save time and resources, SmartFace is unlike other face recognition SDKs requiring biometric expertise to perform advanced demographic analysis. It can be seamlessly integrated into a wide range of applications to carry out real-time face detection and recognition, age/gender detection, and unique visitor counting. With the new update, we’ve even made it more powerful for our users to enjoy the following:

  • Scaling and Performance improvements:
    • Each camera runs in a separate process
    • Matcher is decoupled from camera service
    • Extractor is decoupled from camera service
  • New iFace 3.6 library integrated
  • New license check API endpoint available
  • New keep-alive heartbeat messages available from ZeroMQ
  • Experimental support of USB cameras
  • Some data fetching from web GUI optimizations
  • Several web GUI UX improvements

Related news

Innovatrics SmartFace Adds Face Mask Detection Feature

In response to the curveball thrown by COVID-19, Innovatrics has added Face Mask Detection to SmartFace. The latest version of our facial recognition platform can now detect if someone is wearing a face mask or not, check if it is properly worn, and make sure the nose is fully covered. Especially beneficial in automated access control, this feature... Read more


Digital Onboarding Game Changer: On-Device Passive Facial Liveness Detection

The added biometric layer rivals the speed and accuracy of its server-based counterpart, improving security and user experience during the digital onboarding process. Innovatrics has developed the world’s first single frame passive facial liveness detection algorithm, capable of accurately verifying liveness of an individual performed entirely on-device using a standard selfie camera. Built upon machine... Read more


This site uses cookies to provide services, personalise ads and to analyse traffic. You consent to our cookies if you continue to use our website. More info

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.