Computer Vision
Abrash went on to describe the commonality in computer vision technologies that VR and AR devices will enjoy in the future, as both types of hardware will be capable of rendering convincing mixed reality, where real-world objects and environments are shared with virtual ones. The approaches to this challenge are different; VR headsets of the future will be able to scan and render the real world in the virtual environment in real time, Abrash says (he believes this will be workable within four years), whereas AR glasses ‘overlay’ a rendered layer to the real world seen through a transparent display.
While he thinks that VR will be the “best [form of] mixed reality for a long time,” he also expects the two technologies to eventually share the same underlying interface, and share the same developer environment and tools, with apps and environments working seamlessly across both types of headset.
Virtual Humans
Finally, he spoke about convincing virtual humans, and the uncanny valley that, until recently, seemed almost impossible to overcome. Abrash highlighted a machine learning-based approach called Codec Avatars, running real footage of a talking head next to a remarkably accurate reconstruction. While still in its early stages, he believes that this could revolutionise virtual communication and collaboration. “I’m not betting on having convincingly human avatars within four years,” he said, “but I’m no longer betting against it either.”
“Four years from now, VR is going to jump to the next level,” he surmised. “And that’s just the start. Every area will continue to improve, and virtual humans and likely haptic hands will be along before too long. In short, as far as I can see—and I can see pretty far—the future of VR couldn’t be brighter.”
While Abrash spent his time talking about underlying technologies and capabilities, we recently explored 10 specific projects that have us excited about the future of AR and VR.