Visionaire Docker Snapshot is a containerized Face Recognition to be deployed on top of Docker Runtime

Fast Face Identification

Process Rapidly detects and matches faces from two photos.

Accuracy

Our face recognition achieves 99.83% accuracy using Labelled Faces in the Wild (LFW) dataset. In addition, on our own testing dataset, the system manages to achieve a false non match rate of 17% and a false match rate of 0%.

Tolerance to Face Posture

Assures to correctly recognize a face with a particular condition like minor tilt, nod, age, and minor attributes like glasses and scarf.

Identification Capability

Our face match can be used as a verification system (1-to-N comparison).

Architecture