For the NBA and the largest law enforcement agency in New York, Suspect Tech’s computer vision is shiny and new. For Suspect Tech founder Jacob Sniff ’13, computer vision is something he has been working on for nearly ten years.
Sniff’s experience in computer vision began while an undergrad at Princeton, working in biological imaging research in Professor Thomas Gregor’s quantitative fruit fly lab. Entrepreneurship then came along as he and his brother Matthew founded Photorankr (later renamed Coversplash), an integrated social network, marketplace and portfolio platform for photographers to share, rank and comment on work. Photorankr participated in the Keller Center’s eLab Accelerator in 2013 and generated a lot of interest. At its height, Photorankr had 10,000 users according to Sniff, and members of the founding team later moved on to Instagram and Vine.
Sniff then returned to the computer vision space after graduation, founding Suspect Technologies and developing a facial redaction product for companies that produce body cameras. Law enforcement agencies have been using Suspect Tech’s facial redaction technology to blur the faces of victims and other individuals appearing in publicly available police video to protect their privacy. This process saves law enforcement agencies time and money – redacting faces manually for a 10-minute video can take several hours, but with Suspect Tech, the process can be reduced to as little as 10-20 minutes in some cases.
Sniff has been expanding Suspect Tech’s presence in the facial recognition space, with this technology allowing for law enforcement agencies to quickly identify people in real-time or during post-processing. This approach is especially useful across large areas such as university campuses or crowded events like graduations. “Looking for a person on a bike over 2,000 cameras throughout a university, it’s humanly impossible to do it quickly. You’d have to spend hours,” said Sniff. Evaluated against public databases, Suspect’s technology can deliver state-of-the-art level accuracy, at or near 99 percent true positive for such tasks.
This led to the technology landing in an unlikely place: the locker room of the Dallas Mavericks. Dallas owner Mark Cuban, who had led an investment in Suspect Tech 12 months earlier, installed the facial recognition technology in the Mavs’ locker room. As individual players enter the locker room, their faces are scanned and they are greeted with customized messages and calendars. According to Sniff, “it was a good first deployment of the facial engine.”
Suspect Tech was also part of the second cohort of Princeton Alumni Entrepreneurs Fund portfolio companies. “AEF was very impactful for us. Mark Cuban invested, in part, because we had Princeton willing to invest,” said Sniff. “It’s very hard to find true seed-stage funders, but AEF is one, so it was very helpful in that regard. I’ve always appreciated the Princeton network around entrepreneurship.”
Suspect Tech is working with several universities on beta projects, and with the potential to be a key technology in preventing a future Parkland incident, Sniff is working to position Suspect Tech’s facial recognition solutions front and center. “We have good traction and a good revenue stream with our privacy product with law enforcement, but the facial recognition market with universities, I think it could really take off. We’re going to push forward and see where that goes,” noted Sniff. “We’re trying to raise more funding and see where it goes. We’re a pretty successful company in the AEF portfolio, we have good revenue, but I’m trying to push it farther.”
Visit Suspect Tech’s website at suspecttech.com.