momentas

Inspiration

A photograph captures a moment, but it misses the stories around it. We wanted to help families and friends share those memories and see them through each other’s eyes.

What it does

momentas turns memories into lifelike images. Start with words or a photograph, visualize what happened outside the frame, add stories to meaningful details, and invite someone to share their perspective.

How we built it

We built momentas with Next.js, React, Supabase, and Vercel. Meta’s Muse Image creates the visuals, Muse Spark helps organize memories and connect perspectives, Muse Voice Transcribe captures spoken stories, and SAM outlines objects.

Challenges we ran into

Preserving personal details, selecting small objects accurately, and making voice recording feel natural were challenging. We also needed to clearly distinguish original photographs from AI interpretations.

Accomplishments that we're proud of

We built a working experience that goes beyond image generation: people can share a memory, explore its details, and start a real conversation about what they remember differently.

What we learned

The most meaningful use of AI was helping people express themselves. Small design choices—like keeping the original photograph visible—made the experience clearer and more trustworthy.

What's next for momentas

We want to improve visual accuracy, create shared family memory collections, and make it easier to preserve stories across generations.

Built With

• react