Stable Diffusion is working on generative video
Stability.AI have released the first version of their new generative video software, imaginatively called Stability Video Diffusion. The new program will let you create a video from one still image.
At the moment, it’s only available as a research preview. However, that is likely to change in the near future. There are two image-to-video models capable of generating 14 and 25 frames at different frame rates between 3 and 30 frames per second.
“This state-of-the-art generative AI video model represents a significant step in our journey toward creating models for everyone of every type,” the company writes on its website.
The company is boasting that its version is far superior to other image-to-video generators. However, the resolution is only 576 × 1024px. That’s not really going to cut it for much.
Additionally, it has other shortcomings. The generator is not able to do anything photo-realistic yet, videos only last up to 4 seconds, there’s very little camera motion, and it can’t do text or faces yet. Presumably, hands are also a fail judging by other AI generative software.
The key word here is “yet”. As we have seen with other generative AI, it will probably develop extremely rapidly over the next few months.
Stability.AI has been somewhat cagey about where it sourced its data for this new venture. It has stated that it used videos that were publicly available for research purposes.
Alex Baker is a portrait and lifestyle driven photographer based in Valencia, Spain. She works on a range of projects from commercial to fine art and has had work featured in publications such as The Daily Mail, Conde Nast Traveller and El Mundo, and has exhibited work across Europe