Active Liveness Check
Active liveness check requires the users to perform certain actions, such as turning the head, smiling, or blinking, to prove that they are alive.
Agent process is one of the formats that represents an event occurring within a time period. Each process contains a set of data on human recognition and tracking:
- Process ID
- Process time interval (start and end of the process)
- Process type (track, human, face, body, emotion etc.)
- Detection object with face attributes (object is a human in the camera’s field of view)
- Best shot and an array of all frames received until the person left the camera's field of view.
Bbox (Bounding box) is the rectangle that represents face bounds in the image. Bbox coordinates are calculated relative to the coordinates of the original image.
Biometric Face Template
Biometric face template is a unique set of biometric features extracted from a face image. Templates are used to compare two face images and to determine a degree of their similarity. A biometric face template has the following key characteristics:
- It does not contain personal data
- It cannot be used to restore a face image
- It can be serialized and saved to a file, database, or sent over a network
- It can be indexed. That helps accelerate face template matching by using a special index for face template batch.
Computer Vision Model
Computer vision model is a mathematical model that is created as a result of neural network training and used to process and analyze images and video.
Crowd analytics is the process of collecting and analyzing data from groups of people in public spaces, such as shopping malls, sporting events, transportation hubs etc. This data can include information about foot traffic, demographics and behavior patterns. Crowd analytics can help businesses and organizations understand how people move through and interact with their spaces, identify areas for improvement and make data-driven decisions to enhance the customer experience.
Detector is a libfacerec library algorithm that applies neural networks to detect faces in images, video streams and video files.
Digital signage is the process of broadcasting digital content to multiple advertising screens, and collecting and analyzing information about advertising audience. It is often used in public spaces, such as airports, malls and restaurants, to customize content based on viewing statistics, audience demographics and behavior that potentially increases brand awareness and drives sales.
Distance between compared vectors of biometric face templates. The lower the value, the higher the degree of similarity.
Face crop is an image cropped according to the calculated coordinates of the face bbox.
Face identification is the process of matching a biometric template of a given face image to an existing database of biometric face templates (face search in the database).
Face landmarks are anthropometric points of the human face.
Face normalization refers to the rotation of a non-frontal face to a frontal position. It is needed for better handling of face recognition and other operations with detected faces.
Facial Recognition Model
Face recognition model is a computer vision model that is used to recognize faces in images, video streams and video files.
Facial Recognition Method
Face recognition method is a version of face recognition model (e.g. 12v100, 10v300 etc.).
Face verification is the process of matching two biometric face templates to determine the degree of their similarity.
False Acceptance Rate (FAR)
False acceptance rate shows the system resistance to false acceptance errors. Such an error occurs when the biometric system recognizes a new face as previously detected one, i.e. images of different people are mistaken for images of the same person. This rate is measured by the number of false-acceptance recognitions divided by the total number of recognition attempts.
False Rejection Rate (FRR)
False rejection occurs when a system fails to recognize a previously detected face, i.e. two images of the same person are mistaken for images of different people. The rate shows the percentage of recognition attempts with false rejection result.
Fitter is a special algorithm of the libfacerec library, that positions a set of face landmarks with 2D/3D coordinates linked to a specific detected face.
Head Rotation Angles
There are three head rotation angles:
- Yaw: rotation around vertical Y-axis
- Pitch: rotation around horizontal Z-axis
- Roll: rotation around horizontal X-axis.
Iris landmarks are 40 points of the eyes (pupils and eyelids).
Liveness estimation is a process of determining whether a face, detected in the image or video, is real or fake.
Neural network is a machine learning algorithm modeled after the structure and function of human brain and used in a wide range of applications, including computer vision.
Short time identification (STI) is used to recognize a person who was in the frame some time ago, even if this person is not in the database, or identification is disabled. With STI enabled a person who was tracked, lost, and then tracked again within, for example, one minute, will be recognized as the same person.
Start position is the fixed position of the face in the image. You can set the coordinates of start position if you are sure that there is a face in the given area of the image. The image with the marked start position goes to fake detector, that sends the image directly to the fitter. It is assumed that the image already has a face, which means that you can immediately proceed to the fitting of face landmarks.
True Acceptance Rate (TAR)
Defined as 1 – FRR. True acceptance rate represents the probability, with which the biometric system is able to positively match two images that belong to the same person.
Tracker is an algorithm of the libfacerec library that allows tracking face positions from frame to frame.