The march toward ubiquitous AR
Excitement about augmented reality (AR) hit a fever pitch just a few years ago, with venture capitalists investing billions and crowdfunding campaigns raising millions more on the belief that the technology would transform the world almost overnight.
But while some believe AR has entered the “trough of disillusionment” portion of the technology hype cycle, three AR industry leaders speaking at a gathering of developers argued that it still shows immense potential to eventually reshape our everyday lives.
In a pre-panel presentation, Tilt Five CEO and castAR founder Jeri Ellsworth pointed to the financial challenges facing AR developers but said she was still bullish on the potential for innovation labs and scrappy startups to make a real difference even with the loss of big-money backing.
“I think a lot of innovation comes when things aren’t the most ideal for startups, and people that can push through those tough times actually produce better products in the end,” she said.
Ellsworth believes that hype cycle charts predict absolutely nothing. When outside investments pour in, some companies hit it big while others fail, and it’s the same when those investments slow to a trickle. In other words, there will always be successes and failures, but one may never know when a new product will prove transformative.
Prerequisites for success
In the meantime, Ellsworth suggested that for AR to become ubiquitous it will need to meet four prerequisites: It must find useful vertical markets with use cases that meaningfully improve how specific things get done. It also needs to find an acceptable form factor for each of these markets and to be embraced by end-users. Beyond that, AR just needs time — time for the technology to advance so that more problems are met with solutions, and time for developers to tackle the challenges of AR from different directions.
“There’s no silver bullet out there,” she said. “There’s been this idea that the next pair of glasses coming out by XYZ company is going to be the silver bullet. That’s not going to happen. There’s too much to be done.
“We’ve got to push at all sides,” she added, and like the chaotic, messy evolution of other technologies like mobile phones, or input devices for game consoles, eventually enough ideas will be imagined, tested, improved, and thrown out that we’ll have something that works well enough to be adopted at a mass scale.
In the nearer term, however, “there’s nothing that’s gonna be so good that everyone quits computing on their phones or their computers and just switches tomorrow to some near-eye display or some crazy contact lens,” Ellsworth said.
Srini Venkataramani, Samsung NEXT’s Project Whare senior product manager, noted afterward in a panel discussion that AR is in a much better place right now than the hype cycle would suggest.
Indeed, the latest AR technology offers a multitude of potential and exciting paths forward, and the panelists agreed there’s lots of money to be made in the short-term exploring the niches of AR — those vertical markets Ellsworth described as the first pre-requisite for mass adoption.
Beyond the hype
Ryan Hickman, product manager for Google Robotics, suggested that the hype cycle revolves around “too many people expecting one singular product that does a bazillion things. It has my location, real-time 3D maps, it understands every object in the scene, it has voice recognition and an AI behind it — this is too much. Technology never launches that way.”
The original iPhone launch back in 2007, Hickman argued, was boring in hindsight. Its big features were a web browser, iTunes Store integration, and a touchscreen.
“If Steve Jobs had been on stage in 2007 and said like ‘yeah, you’ll pull out your phone and you’ll call a random stranger to come and pick you up in their car,'” Hickman said, “you’d be like ‘are you crazy!?’ So we need baby steps, and I think we need to go back to some core principles.”
Moving forward requires attention to big-picture business and technology considerations regarding data usage and collection — especially the emergence of point cloud mapping for better integration of virtual objects with the real world.
Hickman said he expects there to be a battle for mapping data, but that it’s unlikely to be a case of winner takes all. And he thinks what really matters is that whatever data is gathered is made interoperable — with the idea being that both mapping data and the labeling of that data for AI training will be shared.
“AR right now is still stuck at this sort of like trying to understand the geometry phase,” he said. “But we’re very close to I’m going to need to recognize a chair and a person and all these other things. We’re going to need real-time AIs on our mobile devices. I certainly hope we don’t have a million AR app companies trying to train up chair recognizers. Some sharing has to happen.”
For Venkataramani, the natural flow will go towards open mapping and data sharing, but not before commercial AR companies emerge to establish the infrastructure — kind of like how, as Ellsworth pointed out, Microsoft Encarta paved the way for Wikipedia.
Making AR matter in our lives
As this new augmented reality ecosystem evolves, the panel suggested that AR’s primary obstacle to ubiquity is not technology — important though that may be. What it needs most of all is to matter: To improve our lives, even if only in some very specific way.
Venkataramani suggested that an app with popularity to rival Pokemon Go that offers more utility to our lives could easily become the catalyst that leads to convergence of design principles and then eventually to mass adoption of AR-capable headsets or glasses.
He added that current AR technology can now support multi-user experiences, albeit only at a small scale, and that presents a great opportunity to make AR more relevant. “I think one of the problems with the way AR is today is there’s only so many puppets dancing on a table or [other] novelty experiences you can watch,” he said, “before you start feeling like, you know, where’s the real value? What am I gonna do with AR that’s part of my everyday life?”
But once users move into multi-user AR apps, he suggested, they can start to experience the technology with family and friends or at an event, “and when that happens it just becomes part of your daily lifestyle.”
In the meantime, AR developers are grappling with challenges as fundamental as how to communicate the strengths — or just the basic concepts — of AR to end users. “I saw a tweet from Rony [Abovitz, CEO] at Magic Leap I thought was pretty spot-on,” Ellsworth said: “Trying to sell AR is like trying to sell TVs on the radio.”
Learn more by watching the presentation.