AI veteran releases 'Hollywood-grade' AI text-to-video tool Odyssey

AI Video & Visuals

Free man recording video stock photo

Two self-driving car veterans have founded artificial intelligence startup Odyssey, which aims to develop “Hollywood-caliber video tools” for professional filmmakers and animators.

“We're building a Hollywood-grade visual AI that can generate and direct beautiful landscapes, characters, lighting and movement,” Odyssey CEO and co-founder Oliver Cameron said on X (formerly Twitter).

Cameron and co-founder Jeff Hawke both have experience building AI at their previous companies, Voyage and Wayve, respectively.

Odyssey predicts it will give companies more control over the output of their AI models, with co-founder Hawk saying, “We believe this technology is the next big thing after text-to-video.”

Odyssey promises to disrupt the AI-powered video generation tool market by giving users full creative control over the final output. “We see this technology as the next step beyond text-to-video,” says co-founder Hawke.

In the X thread, Hawk said Odyssey gives users fine control over the output of their AI models, arguing that if OpenAI's ChatGPT gave users outputs that they couldn't edit, users wouldn't use it.

“But that's how text-to-video works today: the output of the model is a fully baked video file that cannot be edited,” the co-founders added, noting that if the output variables are wrong, users have to regenerate them from scratch.

Odyssey overcomes this limitation by going beyond a single text-to-video model and instead training models to generate “high-quality geometry, photorealistic materials, stunning lighting, and controllable motion.”

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *