A startup is coaching an AI system that it claims will allow creators to generate cinematic worlds, with full management over the surroundings, characters, lighting, and movement. How? By having people strap cameras to their backs and hike world wide.
Odyssey, based by self-driving pioneers Oliver Cameron and Jeff Hawke (Cameron was beforehand the VP of product at Cruise), says it’s created an “advanced camera capture system” that may accumulate information nearly wherever an individual can attain. Weighing about 25 kilos, the system packs six cameras, two lidar sensors, and an inertial measurement unit.
Bearing a resemblance to Google’s Road View Trekker, the system can seize its environment in “3.5K resolution” and 360 levels, with “physics-accurate” depth info metadata connected.
So what’s the purpose? Properly, Odyssey says it’s taking information from the system and feeding it by way of algorithms to “capture the fine details that make up our world.” Primarily, the corporate’s producing digital reconstructions of real-world scenes a la Meta’s Hyperscape undertaking — scenes with forests, caves, trails, seashores, glaciers, parks, buildings, and so forth.
Now, it’s not completely clear how these reconstructions will translate to higher generative instruments for creatives. Cameron and Hawke have beforehand stated that Odyssey has developed a number of generative AI fashions that create layers of visible element, together with object geometry, lighting, and movement, after which mix these right into a single digital “world” to create desired scenes.
Even the perfect “world models” at the moment have limitations, nonetheless — and Odyssey doesn’t declare to have solved all these. However, it’s securing money to forge forward.
Odyssey at the moment introduced that it raised $18 million in a Collection A funding spherical led by EQT Ventures with participation from GV and Air Road Capital. The brand new cash, which brings the corporate’s whole raised to $27 million, might be put towards scaling up Odyssey’s information assortment operations in California.
Odyssey plans to broaden its information assortment to different states and international locations sooner or later — with privateness protections in place, one would hope. (Google’s Road View crew, for one, has discovered itself within the crosshairs of regulators for capturing photographs of public locations that violated bystanders’ privateness.)
“We think it will be impossible for generative models to generate Hollywood-grade worlds that feel alive without training on a vast volume of rich, multimodal real-world 3D data,” the corporate wrote in a publish on its weblog. “We believe an advanced generative world-building model will unlock a better way to create film, games, and more.”
