by Robert Fox
Developed by intelligence agencies in the 1960's, facial recognition technology has, for the better part of 50 years, been what you might call "under development".
We might have seen it in the movies over the years - something out of a slightly dystopian future, featured in hi-tech thrillers like Minority Report.
At the beginning of the 2010's, that began to change, and facial recognition, with the evolution of smart technology, better cameras, and more powerful computers, started to become more feasible to implement into our current technologies.
Now, in 2020, the broader use facial recognition technology has increased significantly, putting some ill at ease, while others feel comforted by this progress.
On the one hand, the increasing popularity of facial recognition brings up many concerns for advocates of privacy, who might expect the technology to run rampant, well beyond the control of its makers.
One might imagine a powerful global mega-corp developing the facial recognition equivalent of the Eye of Sauron. Yikes!
As it stands, some populations already have been placed under watch, presumably for their own good, but also to their general dismay.
Countries like China and England have been rolling out facial recognition programs that monitor their citizens via strategically placed cameras, and there has been some backlash by privacy advocates. At the same time, the advantages here for crime-fighting are hard to ignore.
As such, facial recognition, as a concept, has managed to shed some - but far from all - of the stigma around it, in terms of being an Orwellian "Big Brother".
The fact is, facial recognition is simply another form of biometric security, which we have talked about at length on this website, and which can help to add security to homes and / or businesses if used correctly, providing an effective solution to long-standing access issues around homes and businesses.
Indeed, many forward-thinking companies have seen facial recognition's great potential, and are way ahead of the curve when it comes to its development and implementation into everyday society.
For example, companies like Luxand have been working specifically with facial recognition software for over a decade, and have made great strides towards making such software safely available to anyone who might see a need for it, using API's and SDK's.
For those not familiar, here's a quick look at what Luxand's software can do in terms of mapping out a face:
We are pleased here to get a chance to speak to one of Luxand's reps, Ian Wittig, about their company. We find out how their product, FaceSDK, works, and how their software might apply to some of the things this website likes to focus on, such as home and business security products, and ways for people to increase their overall safety.
Please enjoy our little chat!
Q: Who are you and what do you do?
Ian Wittig, Account Executive at Luxand. I mostly meet with sales and dev teams to plan facial recognition apps from both perspectives (what to build and how to build it).
Q: What does Luxand do / make and how long has the company been in business?
We've been making facial recognition software development kits (SDKs) and application programming interfaces (APIs) for 11 years.
Think of it like a toolbox and all the tools are face rec. functions (detect face, match face, detect eye openess, etc.).
That toolbox is called FaceSDK. You can put FaceSDK inside virtually anything that has or connects to a CPU to give that device access to all of FaceSDK's face rec. tools.
Q: Has Luxand always been a facial recognition company?
Q: What defines good facial recognition software, in your opinion?
Performance and versatility. Performance meaning it works fast and reliably, and versatility meaning the toolbox has tons of different, useful tools in it that will work in many environments.
Q: Is facial recognition a form of biometrics?
Yep. Your face is uniquely yours.
Q: Is facial recognition more effective than, say, a thumb print?
Your thumbprint and your face are both uniquely yours, so they're equal in that sense. In practice what's better is case by case.
The fundamental difference is that you have to physically touch something for print scan, but your face can be detected remotely.
Q: Your company offers facial recognition SDK (FaceSDK), can you describe what that is and how much it costs?
FaceSDK is our flagship facial recognition toolbox. You program what tools to use and how to use them by writing a script in whatever programming language you like.
For home use a Developer License is $99 per month. It's good for one dev on one device for development purposes (or just for fun if you're a home user).
Making commercial apps requires separate licenses depending on deployment specs.
Q: In order to use Luxand software / SDK, do you have to be a developer?
Not specifically. If you have basic experience with a supported language and/or APIs/SDKs you'll find it very user-friendly.
If you know how to compile apps from source code you can easily stand up a proof of concept for a commercial app in <1 workday.
If you have zero programming experience (and aren't interested in learning) you can use our Windows Demo Kit.
PS: FaceSDK has been used in education for intro to APIs/SDKs.
Q: Who would be able to use Luxand Facial Recognition technology, as an end user, and how how could they put it to use?
You can get away with using the Live Recognition and Photo Demo in the Windows Demo Kit for a lot of one-off cases that we're asked about frequently.
Live Recognition runs face detection/match in real time over a webcam. Photo Demo does the same for image files. You can use these two apps to compare faces you find online and see their % match, for example.
People also do this with old photo albums where they're not sure who everyone is (Live Recognition is especially neat for this because it has an age detection toggle. You can just hold the photos up to your webcam for scanning.).
Q: Can you name any notable companies who have put Luxand's SDK to use?
Universal Studios, P&G, Ford, LG, Diebold, Boeing… The list goes on.
Q: Is there a demo of this software that people can see?
See Demos on the FaceSDK download page. Also See our Mirror Reality SDK for demos/videos showing how you can use face rec. for augmented reality.
Q: What makes Luxand hypothetically better than a similar company who might produce this type of product?
There are lots of facial recognition companies with good software. There are a lot fewer who have good software, and first-class customer service (our Account Managers are so dedicated!), and a >10-year track record of being at the forefront continuously and reliably.
Q: What types of home use applications could Luxand have?
Home security, smart doorbells, keyless entry, remote access control, automated sorting/editing photos, ID verification (is it the same face on all social media profiles?), even walk-by analytics.
Q: Do you think Luxand can be useful to the home security market?
Certainly, security is one of our main verticals.
Q: How might Luxand be implemented into a home setting?
Just install on a computer and you can start feeding images and video feeds and programming what you want to do with them.
Q: Any up and coming endeavors Luxand is embarking upon that you could share?
We're focusing on updates for HR and security use cases, like time/attendance, ID verification and visitor analytics.
Specifically we're focusing on liveness detection this year. (You can detect spoofs with FSDK presently, but we want to make it easier to do.)
Thanks for reading our Q&A with Ian Wittig of Luxand on the topic of facial recognition SDK's! If you have any experiences with facial recognition software, we'd love to hear from you. Also, if you have any questions, leave them below! Cheerio!
About Robert Fox
Rob Fox is a former hydro worker who used to teach self defence in Miami for 10 years. He's currently enjoying his retirement, playing cribbage and golf with his buddies, locksmithing and home security in his spare time. Rob is an avid reader, and has even written a few books on the subject of self defence.