Recognition Technologies Still Not Perfected, Media Summit Told
Gesture, voice and facial recognition technologies have come a long way in the past couple of years, but they still have a long way to go, content and ad industry executives told the Digital Hollywood Media Summit in New York Wednesday.
Sign up for a free preview to unlock the rest of this article
Timely, relevant coverage of court proceedings and agency rulings involving tariffs, classification, valuation, origin and antidumping and countervailing duties. Each day, Trade Law Daily subscribers receive a daily headline email, in-depth PDF edition and access to all relevant documents via our trade law source document library and website.
"Between touch, gesture and now voice” recognition technology, there are now “all sorts of ways of getting kids” involved in content “without that barrier of figuring out the interface” of a device, said Miles Ludwig, Sesame Workshop vice president-digital media.
Panel moderator Ted Cohen, managing partner at consulting firm TAG Strategic, saw Samsung demonstrate TV gesture, voice and facial recognition two years ago and “something would go wrong” during each demonstration, he said. Samsung blamed the problems on there being too many gestures, voices and faces in the room, Cohen said. There’s no problem when there is only one person in the room using the technology, but it becomes a problem when there are at least three people in a room watching a football game and “we start yelling things out,” he said.
The Xbox One Kinect sensor’s gesture, voice and facial recognition are “light years beyond” the Kinect for the Xbox 360, said Ludwig. The new sensor’s developers focused on three gestures -- throwing, waving and jumping -- and made sure to “refine the hell out of those gestures so that the Kinect system can read those gestures” by users effectively, he said. “By focusing on that, they got it to work” 90-95 percent of the time, he said.
But 90-95 percent accuracy isn’t good enough, said Ben Huh, CEO of humor website Cheezburger. The “threshold for accuracy … needs to be like 99.9 percent,” he said. “Ninety-five percent sounds good, but if you're at your home and you're giving the system 20 commands and one of the 20 doesn’t work,” the user is likely to think that the system “sucks,” he said. Huh agreed that the Xbox One’s Kinect is “light years ahead” of the 360’s Kinect system, he said. But “it is not prime time” ready yet, he said.
Challenges, meanwhile, remain for augmented reality (AR) to be widely embraced, said Huh. “The problem with AR is it’s a lot of applications, very little reality,” and it will continue to be that way “until we can transform our environment,” he said. AR is just a “stunt” and a “novelty,” complained Ted Mico, chief operating officer at ad company Mirriad.
Barnes & Noble, however, had a positive experience with an AR test at its stores a while ago, said Lee Huang, director-global product management and production. It offered an app in conjunction with Esquire magazine in which visitors who downloaded the app in Barnes & Noble stores could take a picture featuring a virtual version of cover model Brooklyn Decker, he said. It was a “good” experiment that users seemed to enjoy, he said.
Google Glass is also just a novelty for now, said Cohen. “I find it interesting and it’s nice to talk to people about it,” he said. But “so far it doesn’t do anything that my phone doesn’t do better,” he said. The one exception may be the new Refresh app for the device that provides users with instant information about anybody they are planning to meet with, he said.