Odyssey, an artificial intelligence (AI) startup based terminating pace, shared information about its first AI product on Monday. The company unmistakable that it’s development an AI video style that may assemble Hollywood-grade ocular results, similar to OpenAI’s Sora software this is but to be immune by way of the corporate. Odyssey’s co-founder says that the AI style will let customers edit and keep watch over the output at a granular degree, including that the company is operating with a couple of immense language fashions (LLMs) to generate other layers of the output video, which can also be managed one at a time.
How Odyssey’s AI Perceptible Type Works
In a form of posts on X (previously Twitter) Odyssey CEO and Co-Founder Oliver Cameron stated that the AI startup had raised $9 million (more or less Rs. 75.1 crores) in its seed spherical investment led by way of Google Ventures and used to be lately development a device that may ship top quality video that may be customised and edited by way of customers.
Cameron additionally shared information about Odyssey’s AI generation and claims that it used to be designed to generate “Hollywood-grade” video. The chief additionally stated that the startup used to be coaching 4 generative fashions to customers to tug “full control of the core layers of visual storytelling”.
In my view, every style will assist you exactly configure the minutia of your scene.
Mixed, those fashions will generate video or scenes, however precisely as you sought after.
Going additional, our style outputs combine into present pipelines in worth in Hollywood and past. pic.twitter.com/jHZoevLV9n
— Oliver Cameron (@olivercameron) July 8, 2024
Cameron highlighted the disease in present AI text-to-video fashions, which is the dearth of equipment or choices to keep watch over or edit the output. “As a storyteller, you have little ability to direct your environment or characters, or to iterate on the finer details of your shot until it’s just right. More powerful models are required,” he added.
To resolve the disease, the corporate is the usage of a couple of AI fashions that can generate a unmarried layer of the composite video. As according to Cameron, 4 fashions will independently generate geometry, fabrics, lights, and movement. Those 4 layers will likely be generated concurrently in response to a unmarried textual content steered and nearest mixed to give the general video.
The corporate claims that customers will be able to configure every layer one at a time for larger keep watch over over the output. Odyssey can even combine its video outputs into present Hollywood equipment and methods impaired to generate ocular results.
These days, the AI video style is in its early building level. There’s no origination month for the AI style. Then again, the corporate has highlighted that it’s going to accumulation sharing familiar updates about its move. Significantly, Cameron has in the past labored for Cruise and Voyage, two startups running with self-driving automobiles.
Jeff Hawke, the alternative Co-Founder and the CTO of the corporate used to be in the past running because the Vice President of Era at Wayve, an AI company which is growing self sustaining using methods.
