If you have a stash of pictures in your Google Photos library that you didn’t know what to do with, now you do. 

On Wednesday, Google introduced a new Video Remix feature in Google Photos, which lets you instantly transform, relight and stylize your video clips into shareable artwork using AI templates. Video Remix uses Google’s Gemini Omni AI model to filter photos, change the lighting or add a different background to an existing video. The company said it’s a way to “make stylized, imaginative memories from a library of easy-to-use templates.”

Gemini Omni, introduced at Google’s developer conference in May, can create graphics and videos from your prompts. Though there are already several competing generative AI tools at our disposal, this one might help us do something creative with the thousands of photos we have stored away.

Instead of generating something from scratch, it isolates subjects to swap boring backgrounds, applies cinematic lighting to fix bad exposure or transforms your personal clips into 3D animations, watercolors or graphite-looking sketches. It’s specifically designed to be accessible for casual users who can rely on its “one-tap” AI template library to handle the heavy lifting automatically.

Google seems to be positioning Video Remix as a premium AI feature, and cloud processing for these advanced AI video edits is gated behind usage limits. Initially, the feature is available only to adults 18 or older who subscribe to one of its AI plans — Google AI Plus, Pro or Ultra — in select countries such as the US, India, Japan, Mexico and South Korea. 

The feature is located in the Create tab of Google Photos, with several templates to help you get started.

Google’s Gemini AI image and video models are in a race to attract users to paid AI subscriptions. Competitors, including OpenAI, Anthropic and Meta, continue to add AI image and video creation and editing features to their offerings. 

This week, Meta introduced a new image editing tool on the Meta AI app, Instagram and WhatsApp called Muse Image, built to handle more complex requests, create composite photos and edit existing images.