Face recognition algorithm upgrade for Cognitec products
December 12, 2012
Two market-leading face recognition products from Cognitec Systems, FaceVACS-DBScan and FaceVACS-SDK, are now using the B7 engine for premier performance in facial database search and other face recognition applications.
The B7 algorithm shows significant performance increase for recognizing faces in low-resolution images, e.g. low-quality still images extracted from video footage. Cognitec’s research and development team also continued to improve the performance for faces turned to the side, up or down, and for images with noise.
FaceVACS-DBScan 4.8 enables users to compare still images from various sources to those stored in multi-million image databases and instantly view match lists with the most similar identities. The Examiner application within FaceVACS-DBScan provides a powerful tool- and filter set to enhance low-quality images from crime scene photos and surveillance videos, allowing law enforcement staff to achieve better match results when comparing images of offenders against mug shot repositories.
FaceVACS-SDK 8.6, the new version of Cognitec’s comprehensive software development kit, provides all the development tools for face recognition use cases, e.g. enrollment, verification, identification, face tracking, and portrait characteristics measurement. The FaceVACS-SDK has been used by developers and integrators around the world to create state-of-the-art face recognition applications.