Skip to Content

Eagle Eye Networks raises $40 million to invest heavily in AI

Eagle Eye Networks raises $40 million to invest heavily in AI Dean Drako sits down with SSN to talk about how cloud and AI can transform video surveillance

Eagle Eye Networks raises $40 million to invest heavily in AI

AUSTIN, Texas—Cloud video surveillance pioneer Eagle Eye Networks has raised $40 million of Series E funding from venture capital firm Accel, with plans to invest heavily in artificial intelligence.

As Eagle Eye Networks Founder and CEO Dean Drako said in the announcement, there is a “tremendous opportunity to leverage AI and alter the very nature of video surveillance,” noting that “smart cloud video surveillance with appropriate privacy and cybersecurity protections will make business and communities much safer.”

Security Systems News caught up with Drako, via Zoom, to talk about the significance of this round of funding, how the cloud is perfectly suited to leverage AI, and the current and future potential for AI within security and video surveillance.

SSN: Can you talk about the significance of this investment, especially with a VC like Accel?

DRAKO: This is the first time that a tier one VC has made an investment into a video surveillance company … the video surveillance industry, and I am proud to be able to break this ground. This is a really big deal for us, as Accel is very well-known, one of the premier VC firms in Silicon Valley, having funded companies such as Facebook, and Spotify.

We are raising $40 million and we have done a really great job of what I call doing the plumbing for video surveillance — to plumb all of the video to the cloud. We now have proven methods to reliably, robustly and securely get all of the video to the cloud. And now that we have all of that video in the cloud, we can actually start doing all of the AI on it, and that is something no one else can do because nobody else has all of that video in the cloud. Doing AI on the edge is possible, but you just can’t do it as well because you don’t have the compute flexibility and resources [on the edge].

So now we get to execute and start delivering on that dream, or that vision of AI and cloud together, which is what the customer really wants. We literally have ten thousand times more data in the cloud than anyone else. Now, of course, that is all of our customer’s data, but they can run the AI and analytics on it with what we’ve built for them.

SSN: Can you provide a few examples of how customers are using AI today?

DRAKO: We are working with a big car rental company and they want to do things like car tracking, and start to flag for potential problems or anomalies, such as cars gone longer than they should be, for example. This involved doing the LPR component, the face extraction, all of the AI, all done in the cloud, which is much more practical than trying to do something like that on premises across multiple or hundreds of locations.

Another example is the work we are doing with thermal cameras, which are also up in the cloud so you can do elevated temperature detection. And then after detecting an elevated temperature, letting the person know, or their manager know, so there is integration on that level.

Then there is what we call self-trained AI. There are a lot of applications where people want to do their own AI, and many times it is simple stuff — we have hundreds of examples of where somebody wants the system to detect between good and bad. So, for example, messy tables in my restaurant are bad and ones that are clean are good, so the AI can learn what is considered clean or messy and notify a manager or whomever when there is a messy table, or if the lobby gets too crowded, or when cars are parked irregularly or in illegal parking spots, or when someone has fallen down …

This extends out to things like setting up AI to do gun detection. With AI and the training we can do in the cloud, we will be able to do create a gun detection system that is robust, reliable and easy for people to deploy, and that is the key thing. They want to click a button and say, ‘Oh, I just pay a little extra and I get gun detection, great! I don’t have to have IT guys running around at 40 locations installing computers and it doesn't work.’

They want robustness and reliability, which we can deliver in the cloud, but you can’t deliver that on the edge.

SSN: What potential do you see for your channel — systems integrators and security dealers?

DRAKO: I think they are really excited because they have the opportunity to increase revenue by charging the customer for setup and configuration, so the customer can actually see value in that. They see their dealer making it work the way they want it to, and they understand that there is some work involved in that and are willing to pay for it.

So the dealer can add more business in the setup, configuration and management of the system than they might otherwise have not been able to get, so the dealer sees this as a place where they can add even more value.

SSN: What are your short-term and long-term goals with this $40 million in funding?

DRAKO: The part of this that most people don’t think through is that AI, neural networks, machine learning is the new frontier, the Wild West, and [as an industry] we are still figuring it out. Do you use GPUs, or do you not use GPUs, and which GPU should you use? Which algorithm do I use? TensorFlow or no TensorFlow? Do I use Google Cloud or Amazon Cloud, or do I roll out my own cloud? — there are thousands of ways that you can do this.

Eventually, five or 10 years from now, we’ll have it sorted out, but all of this is much easier to deploy in the cloud than on the edge. If I go and deploy GPUs at the edge, two or three years from now you might find that it worked, but now it is out of date and there is something that is much better. Now I have to throw all of that hardware out and get new hardware, but in the cloud world you don’t have to deal with all of that; it is just pay as you go, and as the technology changes, Eagle Eye takes care of all of that for you.

I believe that technology in this AI realm is going to change a lot, but I don’t think we are close to sorting it out as an industry, as a society, as a world — it is changing drastically every six to 12 months. As Nvidia makes innovations, as TensorFlow gets significant upgrades and people make breakthroughs and design and figure stuff out, you don’t want to make a big investment on your own right now to solve problems because whatever you invest in will be out of date.

So Eagle Eye, with our open platform, allows all the third parties to integrate their AI, so we have dozens and dozens of people plugging AI into the platform that our customers can either pay for or buy on a monthly or yearly basis.

We also have our own developments on that front but we are not locking out other folks, so we give the customers lots of choices at a time when there are a lot of choices, and nobody really knows what all the answers are right now when it comes to AI.

So that flexibility is really valuable to the customer and is one of the fundamental premises that Eagle Eye is based on, unlike other vendors who try to lock you into their cameras, their cloud — their every thing.

We work with like 10,000 different cameras and we work with on-premise, or cloud storage, as well as with hundreds of parties who integrate on our platform and integrate their analytics with other features and capabilities. So that is a key part of the Eagle Eye value proposition, and this AI investment is going to enhance that, so that means more capabilities and openness.

SSN: What types of solutions are helping meet the unique demand put on companies during the COVID-19 crisis?

DRAKO: Elevated temperature screening is a big area of focus for us, and we support many elevated temperature thermal cameras that will read people’s forehead temperatures as they walk, or run, by an entrance, and you can measure hundreds of people per minute, which is a far cry from having a little station set up.

So it is a passive screening methodology and we have deployed hundreds of units for customers, such as schools using it to screen students coming in, as well as food manufacturing plants, restaurants, and other businesses that are using it in various locations as a way to get back to work quicker, more easily and safer than before.

Comments

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