VeriLook SDK.

VeriLook SDK capabilities.

• Performance and platforms

  Development platforms

  VeriLook delivers system performance and reliability with live face recognition, simultaneous multiple face recognition, and fast 1-to-1 and 1-to-many matching.

  Available as a software development kit that allows you to develop stand-alone and networked solutions on Microsoft Windows, Linux, macOS, iOS, and Android platforms.

  More details
• Simultaneous processing of several faces

  Simultaneous processing of several faces

  VeriLook 13.0 performs fast and accurate multi-face detection in real-time streaming video and still images. All faces in the current frame are recognized in 0.01 to 0.86 seconds, depending on the selected values for face roll and rotation tolerances, and face detection accuracy. After detection, a set of features is extracted from each face into a template in 0.6 seconds.

  More details
• Recognition of partially closed faces

  Recognition of partially closed faces

  Individuals wearing protective masks (i.e., surgical, fabric or respirators) can be recognized without separate registration.

  More details
• Classification by gender

  Classification by gender

  Additionally, gender can be determined for each person in the image with a predefined degree of accuracy during template extraction.

  More details
• Live face detection

  Live face detection

  A conventional facial recognition system can be fooled by placing a photo in front of the camera. VeriLook is able to prevent such security breaches by determining whether a face in a video stream or individual frame is "live" or a photo. Liveness detection can be performed in a passive mode, when the mechanism evaluates certain facial features, and in an active mode, when the mechanism evaluates the user's reaction to performing actions, such as blinking or head movements.

  More details
• Recognition of emotions

  Recognition of emotions

  VeriLook can be configured to recognize the type of emotion on a person's face. Six main emotions are analyzed: anger, disgust, fear, happiness, sadness and surprise. The confidence value for each of the basic emotions is returned for the face. A higher value for an emotion means that it appears more pronounced on the face.

  More details
• Face attributes

  Face attributes

  VeriLook can be configured to detect specific attributes when extracting a face – smile, open mouth, closed eyes, glasses, dark glasses, beard and moustache.

  More details
• Age estimation

  Age estimation

  VeriLook can further estimate a person's age by analyzing the detected face in the image.

  More details
• Possibility of identification

  Possibility of identification

  VeriLook features can be used in 1-to-1 (verification) mode as well as 1-to-many (identification) mode. The VeriLook 13.0 face pattern matching algorithm can match up to 40,000 faces per second on a PC.

  More details