FaceVACS Technology delivers superior performance for face recognition, age estimation and gender detection
November 16, 2015
A newly released version of Cognitec’s FaceVACS Technology features the next matching algorithms B10 and A16. The much smaller memory footprint of the A16 algorithm enables the development of face recognition applications for mobile devices.
The new release also incorporates an upgraded algorithm for age estimation and gender detection. Cognitec had received the best rating in an independent test for automated age estimation algorithms conducted by the National Institute of Standards and Technology (NIST) in 2014. Since the NIST test, development efforts have further increased the accuracy rates for both age estimation and gender detection.
The development of matching algorithms B10 and A16 focused on improving the recognition performance on difficult image material, in particular facial images with non-frontal poses or strong illumination artifacts, both typical for video surveillance applications. Algorithm optimization also led to better matching results for subjects under the age of 18, and for images, most notably passport photos, with specific types of image noise.
Cognitec’s FaceVACS Technology enables clients worldwide to develop new face recognition applications. It provides a clear and logical API for easy integration in other software programs. Cognitec provides the FaceVACS Technology through customized software development kits, with a set of functions and modules specific to each use case and computing platform. Such particular use cases include image quality checks for biometric photos and image comparisons to large databases, two important practices during ID document issuance processes in many countries.