Face Recognition Technologies
Toshiba Face recognition Software
Face recognition technology is used in a wide variety of applications, including the identification of individuals in video surveillance systems used at public facilities, checking identities at airports and financial institutions, as well as marketing activities to identify personal attributes such as age, gender, or expression.
Toshiba has more than 20 years of experience and a proven track record in face recognition technology, achieving outstanding face recognition performance, high processing speeds, and we have also received excellent feedback in third party evaluations.
In May 2014, NIST*1 reported the results of FRVT 2013 Class C*2 on its website, and in the Executive Summary*3 Toshiba was chosen as one of the leading commercial suppliers.
In another vender test of NIST named "FIVE"*4 which is for evaluation of face in video, Toshiba also received high evaluation.
- *1 NIST: National Institute of Standards and Technology, USA.
-
*2 P. Grother, M. Ngan. Face Recognition Vendor Test (FRVT) Performance of Face Identification Algorithms. NIST
Interagency Report 8009, May 21, 2014.
http://ws680.nist.gov/publication/get_pdf.cfm?pub_id=915761 (PDF). - *3 Executive Summary, NIST Interagency Report 8009, page2.
- *4 P. Grother, G. Quinn, M. Ngan. Face In Video Evaluation (FIVE) Face Recognition of Non-Cooperative Subjects. NIST Interagency Report 8173, March 2017
Toshiba face recognition technology features
Thanks to our worldwide project experience and the over 150 patented technologies we have developed over many years of R&D, Toshiba's face recognition technology can flexibly handle a wide variety of peripheral environments and changes in people. We are contributing to the realization of a safe and secure society based on high-speed processing of vast databases and video surveillance data with world-leading recognition accuracy*5.
- *5 NIST: National Institute of Standards and Technology, USA.
P. Grother, M. Ngan. Face Recognition Vendor Test (FRVT) Performance of Face
Identification Algorithms. NIST Interagency Report 8009, May 21, 2014.
http://ws680.nist.gov/publication/get_pdf.cfm?pub_id=915761 (PDF)
Executive Summary,NIST Interagency Report 8009, page2.
Flexibility in capturing subjects in motion
Excellent recognition performance in actual situations
Toshiba's technology has excellent recognition performance in real-life situations, such as video verification. It is well suited to large-scale video surveillance, and can be used in security applications in crowded places including airports, stadiums, shopping malls, and parking lots.
Expanding possibilities thanks to diverse attribute detection functions
The high-speed, high-accuracy identification function allows secure and rapid personal identification during immigration, during ID issuance background checks (prevention of duplication and fraud), in banking applications, at amusement facilities, and more.
The product is also outfitted with various attribute detection functions, including age and gender, face direction, face coverage level, and smiling level. It can be used for marketing applications such as digital signage and customer analysis.