Skip to Content

Kampr Systems ready to disrupt how industry looks at identity and access control

Kampr Systems ready to disrupt how industry looks at identity and access control Following 15 years as Director of Engineering at Intel, Kahren Arzoumanian tackling access control’s biggest problems

YARMOUTH, Maine—Kahren Arzoumanian is Founder and CTO of Kampr Systems LLC., a new company that introduces biometric access control without credentials. The company's goal: to solve access control's many pain-points, from disassociating from card IDs to solving the security industry's myopic approach to tailgating.

Prior to launching Kampr, Arzoumanian, an engineer with an extensive background in access control, spent 15 years (up until 2020 when he retired) as Director of Engineering at Intel.

“I spent two years in what was called the digital home, then about 10 years in IoT, and the last three years in artificial intelligence,” he told Security Systems News. “With Kampr Systems I wanted to do something that took all of my experiences together and solved as many of the problems and pain-points that exist within access control.”

The following is an exclusive Q&A with Arzoumanian:

SSN: How did the idea for Kampr Systems come about?

Arzoumanian: It started with my brother who has been in the security industry for 30-plus years and runs his own consulting company. You can imagine that we get together and talk about the industry, including what we are each seeing as pain-points, possible wish lists of what each wished that we had.

Back in 2019, he told me that after 30 years we still don’t have a good tailgating solution, meaning one that is robust and doesn’t cost hundreds of thousands of dollars... I started talking with other people and common themes emerged.

First, that we are still dealing with cards and card IDs for identity. Using cards means that when you lose your card you lose your identity.  Card IDs is a more subtle issue. facial readers use card IDs to communicate positive authentication. This means that when you lose your card, you can no longer use your face for authentication.

A second theme that surfaced was that existing solutions are too complicated. There was the desire to have cloud-based edge solutions. After comparative analysis of what was already out there in the market, I found by 2019 that basically access control readers that used facial recognition had a very limited scope: licensed facial recognition software runs on-premises servers and keeps the one-person-at-a-time model, therefore not addressing throughput and tailgating.

The third theme I identified was the solutions out there did not address users’ privacy concerns. This manifested itself through the approaches taken to on-boarding users into the access control system. Facial reader manufacturers provided plug-ins into the access control software that directly handle private information. This seemed strange as readers never have dealt with user information.

And finally, with tailgating being the fourth theme, I threw a fifth challenge:  If we really do a good job at solving those four problems, how can I leverage video to expand from people we are interested in in order to give them access, to looking into unknown persons and all the context that video brings.

SSN: Yes, video, and facial recognition I would imagine?

Arzoumanian: Yes, facial recognition, exactly!

The No. 1 goal was to put all capabilities and decision-making at the edge in a single box. Detecting faces, unknown persons, authenticating, on-boarding, tailgating, liveliness check, communicating with the cloud, communicating with access control panels, all done at the edge. In addition, expanding authentication to up to three persons simultaneously. All this using a 3D camera producing RGB and Infrared channels. I say this because I have heard of the so-called “video chip” usage in cameras used for access control.

On a related note, a lot has been written about usage of facial recognition without user consent. It is important to clarify that the specific concern applies to cameras installed in public places where passersby are not aware. This does not apply to facial recognition readers, as users will opt-in through the on-boarding process and understand that it is one of the most secure methods to identify users.

SSN: You mentioned that the industry has created its only tailgating problem. Please explain.

Arzoumanian: Looking at tailgating I determined with my team that it is comprised of two pieces, and yes, one of them is created by the industry itself. We created tailgating by saying, “Hey, don’t approach the door together, form a line, please approach one at a time and present your card or thumb or face,” for example. We created this due to technical limitations which resulted in people committing tailgating. A majority of people who commit tailgating do it because it is easy and does not have bad intent. Then there is a small percentage of people who have bad intent, so we said we need to solve both, of course, but the first one is the most challenging, and here is why:

If I can now recognize more than one user simultaneously, do I have to create a line anymore? No. With today’s technology, we can do up to three, and with camera technology advancing, that problem is solved too, and soon we will be up to five, or more, and you really don’t need more than that because you must consider the space and the number of faces on the screen, etc.

SSN: What role does AI play in your solution?

Arzoumanian: By adding AI into the equation, we are not only looking for faces, we also have algorithms that check for body features, gait, so even if I hide my face, the system can detect the person. In effect, AI gives us the ability to run specific algorithms that look at different things within the video frames provided by the camera channels simultaneously and with a very high degree of confidence.

SSN: Are you doing away with the access card forever?

Arzoumanian: Yes, Kampr Systems is getting rid of the card and card ID by creating unique digital identity. With secure and unique identity, a simple on-boarding approach that does not involve getting exposed to privacy-related information, and the ability to authenticate multiple users we are addressing gaps that exist today in securing identity, increasing throughput by eliminating the main reason for tailgating, and catching tailgaters with mal intent.

SSN: What is the roadmap moving forward for Kampr? Are you ready to go to market?

Arzoumanian: We are closing beta testing with our first customers. The product is ready, and we are getting set to go to market now that we have proven our concept. We are also in the process of securing equity to help fuel our growth during this important initial launch phase.

As for a roadmap, we are coming out with a tailgate-detector product where no video leaves the camera. This is to address some concerns we’ve heard about privacy, specifically when customers want tailgating detection without persons appearing on video. We are in the final stages of a companion to our reader for datacenters. This product verifies that the person authenticated is the person who enters the premise that is secured. Finally, Kampr Systems plans to work toward standardizing the creation of unique digital identities.

Comments

To comment on this post, please log in to your account or set up an account now.