I just tried Gemini’s new photo-to-video feature, and I’m blown away

Every so often, a new AI feature comes along that doesn’t just feel like an incremental update; it feels like a genuine leap into the future. It’s that “wow” moment where you see the technology do something you didn’t think was quite possible yet. I just had one of those moments.

I just finished trying out the new photo-to-video feature in the Gemini app (and on the web, too), and I’m honestly a little speechless. It’s powered by Google’s incredible Veo 3 model, and the results are nothing short of magical.

Turning any photo into a moving scene

The concept is simple. You upload a static photo into the Gemini app, provide a text prompt describing the motion you want to see, and Gemini gets to work animating it for you. But “animating” doesn’t even begin to do it justice.

I was expecting some simple, maybe slightly janky movement. What I got was something else entirely. The videos it creates are coherent, cinematic, and remarkably realistic. Gemini doesn’t just move the whole picture; it seems to understand the objects, the depth, and the context of the photo. It can add subtle camera pans, make water ripple, cause steam to rise from a coffee cup, or make clouds drift across the sky, all while keeping the rest of the image stable.

For my example, I found a still image of a golfer on the tee box and had Gemini animate him hitting a shot down the fairway. I simply explained clearly what I was after, and Gemini created this video scene from a still image of a guy holding a golf club on a tee box. And the swing mechanics look pretty good!

The level of detail and the realistic motion is what’s so mind-blowing. It’s not just a cool party trick; it’s a powerful new form of creation that feels like it shouldn’t be this easy or this good already. While the swing above is a bit slow, the complex movement of the golf swing is captured quite well, and that’s pretty astounding.

The power of Veo 3

This magic is powered by Veo 3, Google’s state-of-the-art video generation model that we’ve seen produce some incredible results from text prompts alone. Now, by grounding that same technology with a user-provided photo, Google has created a tool that gives you an incredible amount of creative control.

It feels like we’re on the cusp of a new era for content creation, where anyone can add a touch of cinematic flair to their static photos for social media, presentations, or just for fun.

This feature is currently rolling out to Gemini Advanced subscribers, and if you have access, I highly recommend you give it a try. It’s one of those AI advancements that you really have to see to believe. I’m genuinely blown away by what I’ve seen so far and can’t wait to see how this technology continues to evolve.