VeriLook Surveillance 3.0 SDK

Neurotechnology's VeriLook Surveillance 3.0 provides real-time biometric face identification using live video streams from single or multiple high-resolution digital surveillance cameras. The new version not only identifies faces in a crowd, it detects and differentiates people from objects while they are moving through the video frame. It can also detect if a person or object is entering or leaving a predefined boundary area, triggering an alert to the system monitor. This smart monitoring technology can be used for a variety of applications, including retail and commercial areas, entrance monitoring and counting, automated time-attendance systems, transportation, law enforcement and security systems.

VeriLook Surveillance 3.0 includes a robust, dynamic face model that can adapt to visual appearance changes as subjects move across the scene. Based on the latest biometric facial recognition engine from Neurotechnology’s MegaMatcher 5.0 line (VeriLook 5.5), VeriLook Surveillance 3.0 provides even faster face detection, improved face feature determination and emotion detection, including detection of smiles, closed-eyes, open-mouths, clear glasses and dark glasses. VeriLook Surveillance 3.0 matches face images against internal databases, such as authorized personnel or criminal watch lists, after which a VeriLook Surveillance-based application can then lock or unlock a door, put a record into a time/attendance system or immediately trigger a report or alert for recognized (or un-recognized) faces.

In situations when video surveillance is monitoring a very large area or when cameras are in a position such that captured faces are too small for recognition or not visible, VeriLook Surveillance uses moving object detection and classification to identify objects vs. people in the scene. If a moving person or object has crossed a defined border or entered/left a specific monitored area or region, VeriLook Surveillance can send an immediate alert notification and trigger appropriate action if a restricted or sensitive area has been compromised. This ability to detect and track both moving people and objects allows VeriLook Surveillance to be used in a much wider range of applications, such as event security or police applications like the automatic detection of vehicles passing through red lights or stop signs.

VeriLook Surveillance 3.0 continues tracking of subjects even when their faces briefly disappear from the frame or when they are partially blocked by other objects or even other faces (a common problem while tracking multiple faces). The face recognition algorithm enables detection of faces with up to 45 degrees out-of-plane rotation in yaw angle and uses motion prediction models to re-localize faces that have undergone full occlusion, such as when a subject has been fully obstructed by a wall and emerged on the other side. The dynamic face model allows the system to efficiently and reliably track faces in front of complex backgrounds and ensures that subjects can be localized in all video frames, even under strenuous conditions.

VeriLook Surveillance 3.0 enables one computer to process images from several cameras and also provides connections between VeriLook Surveillance units deployed on different computers, synchronizing the databases so that the system works as a whole within the logical network. It enables real-time face detection, extraction and matching by providing parallelization of VeriLook functions for improved performance on multi-core, multi-processor systems.