Cognitec reveals new face matching algorithm
09 July 2018 15:43 GMT

Cognitec Systems has said today that extensive research and development efforts have enabled it to greatly advance the accuracy and speed of its face matching algorithm.

In a statement, the firm said the high-performance algorithm has been integrated into Cognitec’s FaceVACS engine and is now available in new releases of all market-focused Cognitec products. In-house tests show leading-edge performance levels on a variety of large datasets across typical face recognition applications.

These datasets range from passport-style images as found in national ID databases, to mugshot-style images typically used by law enforcement agencies, and facial images turned away from a frontal pose, as recorded by video cameras. Another test shows an accuracy boost for matching low-quality passport-style images to live camera images, the process used for automated border control in eGates.

The new matching algorithm benefits Cognitec’s highly specialized products that are used around the world to compare facial images to those stored in multi-million photo databases, to investigate persons in recorded videos, find known persons in real time, and for verification processes at borders. The FaceVACS engine allows companies to implement face recognition methods in their own products, and is made available via custom software development kits.

Cognitec has written that it is the only company worldwide that has worked exclusively on face recognition technology since its inception in 2002.