Google launched photo-to-video capabilities in Gemini yesterday, allowing users to transform static images into eight-second video clips with AI-generated sound. The feature is available to Google AI Pro and Ultra subscribers in select countries and uses the company’s Veo 3 AI video model. The feature was already available in Flow, Google’s AI filmmaking tool that launched in May, but bringing it to Gemini expands access to a much wider user base. Photo-to-video generation is rolling out on the web today, with mobile users expected to have access by the weekend.
Here’s how it works: Select “Videos” from the tool menu in the prompt box, upload a photo, describe the scene and any audio instructions, and watch as your still image transforms into a dynamic video. Google is currently limiting Veo 3 to producing eight-second clips at 720p in 16:9 landscape format, which means these won’t be optimized for TikTok or Instagram Stories yet.
The numbers tell the story. More than 40 million Veo 3 videos have been generated across the Gemini app and Flow over the last seven weeks. That’s rapid adoption for a premium feature that limits users to three creations per day with no carry-over.
There are some safety features. The videos include a visible watermark to show they are AI-generated and an invisible SynthID digital watermark, too.
The eight-second limit might seem creatively restrictive, but I like the limitation. It forces you to pre-plan your story arcs and to think cinematically about key moments. The discipline is reminiscent of Vine (may it rest in peace).
If you’re new to this, most people start by animating a still image from their photo app. It’s a good place to start – and it will help you get a feel for how this works – but this gets awesome when you use image-to-video to create storyboards using image generators (Imagen, ChatGPT, Midjourney, etc.) then bring them to life. It’s a new workflow that I’m already in love with. You’re going to fall in love with it, too.
Author’s note: This is not a sponsored post. I am the author of this article and it expresses my own opinions. I am not, nor is my company, receiving compensation for it. This work was created with the assistance of various generative AI models.