Crispersoft implements high-end face recognition systems based on AI algorithms. Unlike the other face recognition solutions – we develop systems based on the state of the art approaches in order to provide high precision and real time recognition.
The system could be used in:
For the supervising and monitoring is enough an average IP cameras and processing unit, which could be presented as a cloud computing server or the local computing machine.
We use a special camera reader module in order to stack the obtained frames in the queue for quicker processing. From the queue frame is being redirected and processed by the face detector and feature extractor, the output of feature extractor is a feature vector and it is later stored in the Database. Integrated similarity measure algorithms are used to compare the newly obtained feature vectors with stored ones from the Database, in order to recognize whether a person already exists in the Database or not.
If a person is successfully recognized we can process statistics based on his personality, grant him access etc.
Face recognition and Re-ID methods are used in tracking and combined with our ongoing behavior recognition R&D.