Editorial illustration for OpenCV Creators Unveil CraftStory AI, Powered by Custom-Shot Video Footage
OpenCV Pioneers Launch CraftStory AI for Custom Video Gen
OpenCV founders launch CraftStory AI video startup using proprietary footage
The future of AI video generation isn’t built on YouTube’s blurry leftovers, it’s shot at 120 frames per second, with actors paid for their motion rights. That’s the bet behind CraftStory, a startup launched by the founders of OpenCV, the computer vision library that has quietly powered everything from robot vacuums to autonomous vehicles. Their new model, CraftStory 2.0, sidesteps the internet’s garbage.
Instead of training on thousands of poorly lit, motion-blurred clips scraped from the web, the company hired studios, built high-frame-rate camera rigs, and captured crisp, finger-level detail. The result? A video-to-video system that turns a single still image into a moving portrait by borrowing the gestures of a real actor, who gets a cut every time their movement data is reused.
“You just need high quality data,” says co-founder Ilya Erukhimov. It’s a radical inversion of the AI arms race: less data, better footage, and a business model that respects the people whose bodies make the magic.
Crucially, CraftStory trained its model on proprietary footage rather than relying solely on internet-scraped videos. The company hired studios to shoot actors using high-frame-rate camera systems that capture crisp detail even in fast-moving elements like fingers -- avoiding the motion blur inherent in standard 30-frames-per-second YouTube clips. "What we showed is that you don't need a lot of data and you don't need a lot of training budget to create high quality videos," Erukhimov said.
"You just need high quality data." Model 2.0 currently operates as a video-to-video system: users upload a still image to animate and a "driving video" containing a person whose movements the AI will replicate. CraftStory provides preset driving videos shot with professional actors, who receive revenue shares when their motion data is used, or users can upload their own footage.
CraftStory has drawn a line in the sand. Not with brute compute or petabytes of scraped internet sludge, but with a simple, expensive truth: data quality trumps data quantity. By paying studios and actors to capture clean, high-frame-rate motion, Erukhimov and his team have built a model that doesn’t need to hallucinate fingers.
It saw them clearly from the start. The video-to-video approach is a pragmatic bridge, not a leap into the uncanny valley. Users bring a still image; CraftStory supplies the movement.
And those movements aren’t stolen. They’re licensed, with a revenue share for the performers. It’s a small but crucial moral win in an industry that often forgets the humans behind the training data.
If this scales, the implications ripple outward. Smaller budgets. Fewer compromises.
A future where high-quality video synthesis isn’t the exclusive domain of trillion-dollar labs. The OpenCV founders aren’t trying to beat OpenAI and Google at their own game. They’re rewriting the playbook, one crisp, high-frame-rate frame at a time.
Related Reading
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
From Archives Zoho One adds AI context layer via Zia, serving 75,000+ customers using 22 apps
From Archives Polestar Analytics says winners will turn unstructured data into value
Common Questions Answered
How does CraftStory's video generation approach differ from other AI video tools?
Unlike most AI video generation tools that scrape random internet footage, CraftStory uses custom-shot, high-frame-rate video footage specifically captured by hired studios. This approach allows for crisper detail and better motion capture, particularly in complex movements like hand gestures.
What technology did the OpenCV creators use to improve video generation quality?
CraftStory utilized high-frame-rate camera systems to capture extremely detailed video footage for training their AI model. By shooting actors with specialized cameras that capture more frames per second than standard video, they can avoid typical motion blur issues and create more precise synthetic video content.
What does CraftStory's founder claim about AI video generation and training data?
According to founder Erukhimov, CraftStory demonstrated that high-quality video generation doesn't require massive datasets or enormous training budgets. By strategically shooting proprietary footage and using advanced camera technology, they can create compelling AI-generated videos more efficiently than traditional approaches.
Further Reading
- CraftStory Unveils First AI Model to Create 5-Minute, Studio-Quality Human Videos — PR Newswire
- OpenCV lanza CraftStory: IA de video para enfrentar a OpenAI y Google — Ecosistema Startup
- About us - CraftStory — CraftStory
