KeyLemon S.A. is a Swiss limited company offering convenient, secure and continuous biometric authentication solutions based on face and speaker recognition. To improve robustness to illumination and pose, as well as to provide enhanced security against photo/video spoofing attacks, KeyLemon’s latest face recognition algorithms take full benefit of 3D depth sense cameras by efficiently combining depth, near-infrared and color information. KeyLemon’s advanced technology is available as a multi-platform SDK, online web service and in a unique client-server solution for developers and manufacturers across a wide variety of industries. Used by consumers and enterprises worldwide, KeyLemon can easily be integrated into any device, system or application to improve online transaction authentication, facilitate access to mobile apps and services, and enhance safety features in machinery such as automobiles. For more information, visit KeyLemon.
Is the identification of the person who is in front of the camera by characteristics of their face (face biometrics). Initially, the face recognition system uses a face recognition algorithm to identify human faces in video frame. This algorithm uses an innovative pattern detection algorithm efficient in terms of speed and robustness to illumination. Cameras which include an IR illuminator and receptor are also much more tolerant to variation in lighting. Then the detected faces are scaled, photometrically normalized and decomposed into an efficient digital representation. Using an advanced machine learning technique, the digital representation is transformed into a compact model of identity. Finally, the identity are compared with a facial database containing other identities.
Is the identification of the person who is speaking by characteristics of their voice (voice biometrics) also called speaker recognition. Recognizing the speaker can simplify the task of voice analysis in systems that have been trained on specific person’s voices or it can be used to authenticate or verify the identity of a speaker as part of an authentication process.