Facial Recognition Technology Developed at UCF used to Improve Human Analytics
November 24, 2015
Well-Suited to Address Hospitality, Workforce and Electronic Payment Industries, The Technology Provides Faster and More Precise Photo/Video Facial Recognition
Thomas O’Neal, associate vice president for the Office of Research & Commercialization (left) with Brian Brackeen, Kairos CEO
An innovative, more precise and faster facial recognition technology—compared to the current state of the art—developed through the UCF Center for Research in Computer Vision has been licensed to human analytics company Kairos. The Miami-based startup will incorporate this new computer vision technology within its core facial recognition product designed to serve enterprise customers within a variety of industries including hospitality, workforce and electronic payment industries.
Under the leadership and direction of professor Mubarak Shah, a widely recognized leading expert in computer vision research, the UCF Center for Research in Computer Vision developed the technology to significantly improve the ability to accurately complete facial identification of individuals in photos and video. The video capability is especially exciting for the industry, as traditional technologies have lacked accuracy, are typically cumbersome to operate, as well as expensive to launch and maintain.
“The capability of this facial recognition technology is very exciting to us,” said Brian Brackeen, CEO of Kairos. “Facial recognition technology has been developing over the past few decades, however, video—with its motion and variations in illumination—has long been the challenge. Dr. Shah and his team have developed the right solution that combines accuracy, speed and deployability for the commercial market.”
The technology is based on an algorithm developed by the UCF team to focus on specific facial features that then converts a photo into a checkerboard of patches and extracted tiny snapshots of the most significant facial parts. The computer program than compares the photo against a database of photos, feature by feature, and identifies the match. This is highly attractive within the hospitality, amusement, healthcare and electronic payment industries where organizations find it beneficial to quickly and reliably identify consumers, whether for promotional opportunities or access issues.
In its application for video, the technology provides an end-to-end face recognition system that addresses the difficult problem of identifying a video face track using a large database of still face images. Traditional technologies are based on an expensive frame-by-frame method. Dr. Shah and his team of researchers developed a novel algorithm Mean Sequence SRC that performs video face recognition using a joint optimization leveraging all of the available video data and the knowledge that the face track frames belong to the same individual.
Whereas the defense and security industries feature a number of facial recognition technologies and companies, Kairos is targeting the enterprise market, which it believes is underserved and therefore represents a significant opportunity. Key industries include hospitality, amusement, workforce, gaming, healthcare and electronic payment industries.
“Working with Kairos provides the perfect illustration of how UCF research and UCF-developed technologies are not only brought to fruition but are also applied to the commercial marketplace,” said Narasimha Nagaiah (Raju), licensing associate with UCF’s Office of Technology Transfer, who facilitated the license agreement.
Brackeen, who believes UCF’s computer vision program is one of the top programs of its kind, expressed significant satisfaction in his experience working with UCF’s Office of Technology Transfer.
“In our efforts to identify a leading-edge facial recognition technology, we were quickly able to come across
UCF’s computer vision program and identify Dr. Shah’s research as groundbreaking,” said Brackeen. “The response by the Office of Technology Transfer was exceptional and thorough in its support of our effort to take this technology to the commercial market.”
In addition to working with UCF’s Office of Technology Transfer, Kairos recently received investment support through the Florida Angel Nexus and the Florida Institute for the Commercialization of Public Research. Kairos plans to work closely with UCF and the computer vision program, as the company grows and adds talent.
About Kairos Kairos is a human analytics company designed for innovative, data driven businesses. Its core business offerings provide facial recognition, crowd analytics and emotion analysis solutions. The company provides multi-dimension analysis of human beings by capturing decisive moments in people’s behavior. This intelligence empowers companies with meaningful metrics. Kairos’ mission is to radically change how companies understand people. For more information, visit, www.kairos.com