Certiface antispoofing: available repository!
Certiface AntiSpoofing use oneAPI for fast decode video for perform liveness detection with inference. The system is capable of spotting fake faces and performing anti-face spoofing in face recognition systems. The user could try to hold up a your photo. Maybe they even have a photo or video on their smartphone (obtained in facebook, for example).
Certiface technology was designed to be used in Retail and Banking system to combat fraud and protect honest people by using technology to differentiate between a living person and a recorded video.
Certiface AntiSpoofing use oneAPI and one VPL for fast decode video oneVPL Library enables allows the development of real-time transcoding, decoding and encoding, high-speed direct access to advanced Intel CPU and GPU instructions.
oneAPI unifies and simplifies programming across CPUs and accelerators, delivering developer productivity across architectures and vendors.
Technologies Used: oneAPI Beta 10 and openCV 4.5.0
In the repository https://github.com/cabelo/oneapi-antispoofing, step by step instructions on how to download and compile the system.
Bellow an example running in my machine with OpenSUSE Leap 15.2.
