MWC 2018

Spoof-Proof Facial Recognition? Xperi Showed It’s Possible With FaceSafe

facesafe fotonation xperi mwc18

It’s the first day of Mobile World Congress and we had a hard time deciding where to go first, simply because there were too many exciting things to talk about! Every smartphone maker showed off its babies but we were equally interested in the tech that makes it all possible. After all, the brainiacs who make the software necessary for digital payments and phone unlocking via biometrics are at the intersection of science and magic! With that in mind, we stopped by FotoNation, an Xperi brand, which is taking the lead in depth-based facial recognition. Essentially, we’re talking about a spoof-proof computer vision technology!

This year, at Mobile World Congress, FotoNation has brought forward FaceSafe. As the name suggests, this is tech meant to ensure users that, once they register their face on any mobile device, no one will be able to trick it and log in without their authentication.

You’re probably thinking… that sounds too good to be true, especially if you’ve seen hackers succeeding in their attempts to trick facial recognition systems with pictures before. Or if you’ve read about people managing to trick their friends’ phones to let them in. 

But there’s a good reason why a high resolution photo wouldn’t work to bypass FotoNation’s system. FaceSafe is depth-based facial recognition. This means 2D formats don’t stand a chance against it!

As Geir Skaaden, executive vice president and chief products and services officer at Xperi Corporation, put it, “FotoNation’s FaceSafe solution is an exciting step toward bringing a robust 3D face recognition solution to market, and we are thrilled to showcase the technology at Mobile World Congress,”. He believes “the need for protection and customization becomes more prominent as biometric face recognition continues to impact the mobile market. FaceSafe’s cutting-edge technology is tackling these challenges to provide secure, class-leading user experiences.”

xperi booth mwc18

Indeed, the solution is entirely customizable and has a false acceptance rate of one in one million and a fast execution time that is less than 150 milliseconds! This is a great advantage for businesses as well as consumers; think about it – the more personalization brands can do, the higher the odds are for more of them to adopt such biometric solutions and offer end-users multiple options of devices to choose from. On the other hand, accurate results mean credibility and a “wow” factor for OEMs as well as a sense of security for device owners.

Not to mention, the small memory footprint yields power and system cost savings. This translates to a smaller burden for the processor and ultimately the battery. So, you can actually preserve the performance of your device, keep it alive longer and make it smarter! Win-win situation right there!

FotoNation’ representatives told us that the solution works in diverse light conditions and is not impacted by objects such as sunglasses. Winter, summer, night or day, you get to keep your data safe, make payments securely and open your phone without typing a password or “thumbling” with the device.

Bottom line: we’re moving to an Edge of Things era, where machine learning is at the center of all smart devices and technologies. In the case of FaceSafe, neural network technology is joined by state of the art liveness detection to give life to a computer vision solution that is not tricked by mere pictures. It’s all about depth!

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

To Top