Neurotechnology Advances With New Face Algorithm
Neurotechnology is a provider of high-quality biometric identification technologies of today's generation. It has currently released the so-called VeriLook, which is a face recognition algorithm.
According to Canada Journal, VeriLook is introduced with "a new face verification component" with a more intuitive but simpler interface that is easily integrated and used for authentication tasks like mobile banking transaction verifications.
It is said that with its new algorithm, it has five times higher accuracy in the identification of full frontal faces. Also, it has 10 to 15 times higher accuracy in unrestricted facial recognition.
Dr. Justas Kranauskas, Neurotechnology's project leader, said "We achieved a ten-fold accuracy improvement on faces captured in lower resolution, with complex illumination, expressions and head rotations. This enabled us to offer a new face verification component which greatly simplifies user authentication by face, especially in mobile applications, while also enabling the face recognition algorithm to be used for complex 1:N identification."
Moreover, as reported by Biometric, the advancement of the face recognition algorithm makes it importantly easier to utilize and apply to a more broader range of facial recognition applications, i.e., automated facial image searches in huge databases. By this, it would not need manual review. Besides, it would be able to detect faces accurately in a much faster way.
It is further reported that VeriLook is a part of the new MegaMatcher 9.0 line of biometric software development kits (SDK). Also released along with it are fingerprint, iris, palmprint, face and voiceprint (speaker) identification technologies that could work seamlessly together and can be utilized in any combination for multi-biometric solutions.
In addition, the improvements also span the easy detection of facial attributes that include smile, gender, open mouth, closed eyes, dark glasses and eyeglasses. This new facial landmarks tracking and detection capabilities are more well made in a wider range of facial poses.
Non-compulsory liveness detection determines if the algorithm is looking at a live person or a photograph. "A specialized API simplifies integration into a variety of solutions, and component pricing makes the face verification component economical for large-scale deployments on millions of devices," as further noted by Biometric.
Google Begins Test on Face Authentication Technology: Passwords to be Obsolete by the End of the Year?
Google is working on new cutting edge technology that will probably put an end to the password.
Chinese Researchers Built Realistic Robot Jia Jia That Can Chat With Humans
Scientists in China built a realistic robot hostess called Jia Jia.
iPhone 8 Can Scan A User's Face While Device Is Lying Flat, HomePod Firmware Suggests
Apple's iPhone 8 is coming with an advanced facial recognition feature to give extra security to users. HomePod Firmware announced in their recent update that users can unlock their phone in any angle once their facial data registered to the device.
Emotional Pet Robots Can Boost Health And Happiness Of The Elderly
Pet robots are now being used to provide emotional needs of lonely elderly. The animatronics are endowed with facial recognition skills and emotional responsiveness to better accompany the elderly.
Facezam: Facial Recognition App That Matches Pictures Of Strangers With Their Facebook Profile
Facezam is a facial recognition app that has a 70 percent accuracy rate of tracking down anyone on Facebook just by analyzing a picture of their face.
MORE IN ITECHPOST
Never Before Seen Beauty of Himalayas in the Last 30 Years Captured By The Lens, Thanks To Coronavirus Lockdown!
Different social media sites were filled with some delightful images as more and more residents laid eyes on the entire Himalayan range
NASA Spotted Record-Breaking 1 Million Square Kilometer Rare Hole Above Arctic
The rare hole has opened up in the ozone layer above the Arctic which is set to be the largest of its kind ever detected.
Valorant Won't Kill Fortnite: Tfue Explains Why
A certain twitch streamer by the name of Tfue has just recently tried out Riot Games' Valorant explaining to his followers that the games do not cater to the same audiences.