Due to heightened security concerns, more companies are beefing up their video surveillance systems in lobbies, corridors and stairwells of their offices. But these surveillance systems produce nonstop video that is time-consuming to monitor and review when an incident occurs. And all of that video requires a great deal of network bandwidth to transmit as well as storage space.
That’s where start-up 3VR Security enters, with an appliance that makes it as easy to search surveillance video as it is to type in a query into Google.
3VR sells an intelligent video management system that automatically categorizes surveillance video using biometrics and other analytical techniques and creates metadata about the images. Users can quickly search the metadata to find a particular person on video.
Even more important for network managers, 3VR’s metadata can be easily transmitted without requiring much network bandwidth. The metadata also diminishes the need for long-term storage of all that video.
3VR was founded in 2002 and funded by In-Q-Tel, the investment arm of the CIA, along with venture capital firms Kleiner Perkins and Vantage Point. About 25% of 3VR’s customers are government agencies, with the rest being banks, pharmaceuticals, hotels and other enterprises with security concerns.
The 3VR Intelligent Video Management System has been shipping for a year. The appliance looks like a regular security camera, and it uses biometrics and other analytics to categorize all of the images taken by the system.
“If you walked into the lobby of a building, a normal camera would take 3,000 pictures of you. What our system does is say there is a person walking in this direction. Here is the best shot of this person’s face. We apply a biometric template to the face, and we do this all in real time,” explains Tim Ross, co-founder and executive vice president of 3VR. “Then when you search through this video, you can start with a shot of a person’s face and tell the system to show you anytime that person has been in the building or anyone who has been with this person. The result is mug shots.”
Ross says it is difficult for companies or government agencies to review surveillance data captured by hundreds of cameras at different locations because that requires transmitting all of that video over the network to a central location. With the 3VR system, only the metadata gets transferred over the network.
“We’ve changed the problem because we’re not thinking about coming up with better ways to cache video or compress it. We’re asking why centralize video? By creating a set of metadata about the video, instead of sending a 2M file we’re sending a 10K file,” Ross says. “We can centralize the metadata and send that very easily over a conventional network and leave the video at the edge.”
The 3VR system also allows users to store only the metadata rather than all of the surveillance video, for a significant savings.
3VR systems sell for around $1,200 per camera. Version 5.0 of the system’s software was released in September.
The Sir Francis Drake Hotel in San Francisco purchased 16 3VR cameras, which record video in public areas. David Garon, hotel manager, says he needs to go back and review surveillance video at least once a month.
Garon recently used the system to identify a thief who was breaking into the hotel’s linen closets. He says the 3VR system saved time in the investigation and money because the thief was captured faster and couldn’t steal as many items while the investigation was ongoing.
“Through the 3VR systems, we were able to identify the time the break-ins were happening, who the person was, and catch the person with great quickness as opposed to going through miles and miles of video,” Garon says. “We were able to isolate the face of the person who was doing this. We weren’t wasting time looking at other people…We saved days on that investigation.”
Garon says the 3VR system is easy to learn and use for security personnel, who are not always tech-savvy. So far he hasn’t run into any problems with the system.
“We’re finding all kinds of direct and indirect efficiencies with this system,” Garon says. “Reduced storage is one. We don’t have to store video, which deteriorates over time. Another savings is the ability of the system to erase video in areas that are quiet. We have some corridors that are dormant from 1 a.m. to 4 a.m. Why should we keep that video? We can delete it and increase our storage space. But the man-hours for reviewing and searching video are the big savings.”
