SkyBiometry, a subsidiary of Neurotechnology, a provider of deep-learning-based solutions, robotics and high-precision biometric identification technologies, has announced the release of the new version of its face detection and recognition algorithm.Significant enhancements make the algorithm much more accurate and allow for a wider range of facial attributes in the detection process.The latest face detection algorithm locates many more faces in varied and difficult conditions and can detect multiple head rotation angles up to, and including, a full profile. Compared to the previous version, face recognition accuracy is five times higher when set at a low False Acceptance Rate. Other face processing algorithms within the API have also been updated, including more accurate facial-feature points detection, better face image quality estimation and improved face properties and emotion classifiers.”The new features made available in the latest version of the SkyBiometry facial algorithm were added and augmented based on customer feedback,” said Dr. Justas Kranauskas, head of the biometric research department for SkyBiometry and Neurotechnology. “More accurate face detection and face recognition algorithms continue to make our service unique, and the additional attributes will help our customers expand the scope of their SkyBiometry API applications.”Previous versions of the algorithm were able to determine such attributes as gender, age, smile (or not), glasses, dark glasses, lips (parted/sealed), eyes (open/closed), emotions (neutral, angry, disgusted, scared, happy, sad, surprised), roll, yaw, and facial feature points. Additions to the SkyBiometry algorithm can now establish whether a person is wearing a hat, has a beard or a mustache and determine race/ethnicity. Detected attributes may be used in a wide range of scenarios, from entertainment applications to advanced security and image moderation.The latest facial recognition algorithms achieved high scores in NIST FRVT evaluations:FRVT 2018 – The face identification algorithm ranked among the six most accurate algorithms (out of 40 different vendors) in the largest test, a population of 12 million subjects, and ranked 3rd (out of 40) when face photos were taken up to 18 years after the initial enrollment photo.FRVT Ongoing – The face verification algorithm ranks among the 15 most accurate algorithms (out of 82 different vendors) in the “wild” test, identifying faces culled from various real-world sources.The SkyBiometry API can be used for face detection, recognition, grouping and attributes determination. It simplifies the process of identifying and cataloging images for photographers and others who work with large photo databases. For example, it can be used to build a comprehensive system that searches through image databases to identify individuals in photos, eliminating a great deal of manual effort.The API can also be applied in developing time and attendance software, using facial identification to speed up and automate the time entry process and ensure the accuracy of individuals who are clocking in and clocking out. It can be used as a simple, fast and secure form of identification for user authentication by face. And the SkyBiometry API can also be used in marketing and entertainment applications, for building interactive advertisements or mobile applications or to moderate pictures in social media and dating sites.To start using SkyBiometry, a simple registration is required. Upon registering, the user has access to a free subscription. The subscription includes 5000 monthly calls (requests) and a total of 1000 trained tags (faces) of database capacity.SkyBiometry has three different demos available: face detection, face recognition and face grouping. It enables algorithm testing without any initial programming.The SkyBiometry API is available through SkyBiometry or Neurotechnology. The entire line of products for AI, robotics, object recognition and biometric identification is available through Neurotechnology and from distributors worldwide. For more information, go to http://www.neurotechnology.com.