Could video transformation become more important than video generation?

I tried Google's latest video model, Gemini Omni, using my own footage as input and asked it to transform the scene from warm lighting to a cool blue mood, and then to bright studio lighting.

As someone who loves Nano Banana, I had pretty high expectations for this.

And I have to say, I think it did a surprisingly good job.

The consistency wasn't perfect, but it managed to preserve most of the original scene while changing the mood and lighting.

Being able to explore multiple moods from the same footage could be a powerful tool for storytelling.

Sharing my process in the comments.

What's the first thing you'd try with this?

Process walkthrough:

I built an anonymous corner for the internet. It's called the Fruit Market. 🍊🍓 Every visitor leaves behind a unique fruit that represents them. You choose a fruit, give it a label, add a few pieces of identity, and answer a prompt about something you're carrying, longing for, or wishing someone understood. Then you place it on the shelf. The result is a market full of anonymous human artifacts: A mango who is missing home. A peach who is terrified of graduating. A dragon fruit who is searching for purpose. The idea came from a question: How deeply can we understand each other when names are removed? Built with Figma and Figma Make on my humble phone and tablet. My goal was to create a whimsical, imperfect, yet vulnerable space that feels somewhere between a community bulletin board, a missed-connections page, and a message in a bottle. Please, if you’re willing, leave a piece of yourself at the Fruit Market. <3 Website: thefruitmarket.figma.site LinkedIn Post (feel free to connect!): https://www.linkedin.com/feed/update/urn:li:activity:7471214902842470401/ Figma Community Link (a copy of the website w/o the backend): https://www.figma.com/community/file/1647270469719499321 Made for the Contra x Figma Makeathon.

Wonderful and worth th build