Cognitec exec discusses FaceVACS facial recognition technology

January 5, 2016 - 

Cognitec’s FaceVACS facial recognition technology enables companies and organizations around the world to develop new face recognition applications through an API that can be seamlessly integrated in other software programs.

Available through customized software development kits, FaceVACS delivers numerous functions and modules that are personalized for each computing platform and use case, including image quality checks for biometric photos and image comparisons against large databases.

Cognitec recently added matching algorithms B10 and A16 with its latest version of FaceVACS, which improves the recognition performance on difficult image material.

The new algorithms help to achieve more accurate matching results for subjects under the age of 18, as well as for those images that are hindered by specific types of image noise, such as passports.

BiometricUpdate.com recently had the opportunity to talk to Elke Olberg, Cognitec’s marketing manager, where she discussed the company’s FaceVACS technology, its new algorithms, and how the technology has helped to dispel common misconceptions about facial recognition.

Cognitec recently released a new version of its facial recognition FaceVACS Technology with new matching algorithms B10 and A16. How has this improved the technology?

Elke Olberg: Obviously, with each release of the FaceVACS Technology, we try to improve the overall image comparison performance and therefore the recognition accuracy and speed for all Cognitec products. All market-leading algorithms are now almost perfectly matching high-quality facial images, so we are working on continually improving the recognition performance on difficult image material. Our comparison tests on the current algorithms showed a significant performance improvement for 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 1 8, and for images, most notably passport photos, with specific types of image noise.

Age estimation and gender detection are two major features of Cognitec’s technology. How can these help benefit companies and their end users?

We have been promoting the use of anonymous facial analysis for business intelligence and people flow applications, but many use cases are demanding higher accuracy rates to justify the implementation of this technology in comparison to more conventional methods. As the performance improves, end users will be able to rely on meaningful, consistent data about people count, dwell times, gender and age distribution, and traffic patterns, just to name a few statistics the technology currently measures. On the other hand, searching for a male or female person in a particular age range can also aid criminal investigations and real-time surveillance tasks.

What are some common misconceptions about facial recognition technology among the public and what how can Cognitec’s technology contribute to dispelling them?

People have legitimate concerns that every image of their face, taken by security cameras, phone cameras, immigration desk cameras, passport agency cameras, etc., can all be connected with each other via face recognition and then to other personal data; that such “super searches” will generate a comprehensive surveillance trail for each person. While such powerful, fully connected systems do not exist at the moment, we certainly need to watch out for the systems watching us. Future technological developments and data protection legislation need to ensure that face recognition and other biometric technology is used for the right reasons, in the right place and with the right regulations. As developers of a complex, but ultimately very personal technology, Cognitec must be committed to promote privacy by design, responsible use, and open communication with our partners and the general public.

What are some key trends you have seen developing in 2015 in the biometrics market, specifically in the facial recognition space?

In general, the public has gained enough knowledge about what biometrics are, what they can do, where they can be abused, and which applications make sense. The industry and its associations are coming together in educating all users more clearly about currently deployed biometric technologies and have begun to survey public conception in regards to acceptable, responsible use of biometrics in everyday life. As nations are uniting in the fight against terrorism and in the management of refugee migrations, face recognition technologies can support the immediate demands for heightened border control and public security.

Leave a Comment

comments

About Justin Lee

Justin Lee has been a contributor with Biometric Update since 2014. Previously, he was a staff writer for web hosting magazine and website, theWHIR. For more than a decade, Justin has written for various publications on issues relating to technology, arts and culture, and entertainment. Follow him on Twitter @BiometricJustin.