Google Gemini Can Now Turn Photos Into Videos

It’s now possible to turn photos into videos using Google Gemini’s Veo 3 AI technology.

PetaPixel has already highlighted the crazy, and somewhat scary, things that Veo 3 can do in terms of generating videos from simple text prompts, but now the technology can turn a single still photo into an eight-second video using generative AI, including sound.

“Your imagination is the limit when you create videos with Gemini,” Google promises, alluding to a future where a still frame of a real moment can be turned into something altogether different in seconds, or something that never existed at all can pop into life as a photorealistic video.

As is the case with many of Google’s other Gemini AI tools, the ability to turn a photo into a video is very straightforward. Users enter Gemini, select “Videos” from the tool menu, and then upload the picture they want to animate. Users can describe what they want done to their image in a text box.

“You can get creative by animating everyday objects, bringing your drawings and paintings to life or adding movement to nature scenes,” Google explains. “Once your video is complete, tap the share button or download it to share with friends and family.”

As Engadget notes in its coverage, Gemini’s Veo 3 is currently capped at eight-second 720p video output in a 16:9 landscape format. However, social media formats and higher resolution output options are inevitable.

The photo-to-video capability within Gemini is now available to Google AI Pro and Ultra subscribers in select countries and will be rolled out to additional markets and users later. It is also available now within Google Flow, Gemini’s AI platform aimed specifically at filmmakers, which includes additional cinematic tools and options. Google AI Pro starts at $20 monthly, while the company’s Ultra subscription costs $250 a month.

It’s easy to imagine how badly AI-generated video can go wrong, so Google maintains that it’s committed to safety in the age of AI, including ongoing evaluations of its tools and dedicated teams aimed at addressing problems “before they arise.” All generated videos include a visible watermark and feature Google’s invisible SynthID digital watermark, which Google added to most images adjusted in Pixel’s Magic Editor earlier this year.


Image credits: Google

Discussion