How do I make the first time in a fancy cafe AI video with my own photo?

How to make the first time in a fancy cafe AI video with your own photo

Updated 2026-11-03 · Fancy Cafe AI

Upload one photo of you (no sign-up needed) and the AI casts you as the one perched on the tiny cafe table; add a second photo if a friend should play the guest who walks in. Pay for just that video or use credits, and a 10 or 15-second vertical MP4 is ready a few minutes later: you sip on the table, the guest sits the right way, you hop down as if nothing happened.

The photo — One generation from this photo, shown as it came out.

Step by step

Choose one photo. One subject per photo, sharp and in daylight: people waist up or head to toe facing the camera, pets with the whole body in view. Everyday clothes look most natural; your clothes go into the clip.
Upload it in the box above (no sign-in needed) and pick the length (10 or 15 seconds), quality and format. Tapping the button confirms the photos are allowed: photos of yourself, of adults who agreed, or of your own pets.
Pay for just that video (⁠$9.99⁠, one payment, no subscription), or use credits: a 15-second 720p video uses 60 credits; credits from ⁠$9.99⁠ for 60, or a monthly plan.
Wait 2 to 5 minutes. The video shows up in My videos when it's ready.
Download the MP4 (on iPhone, Save to Photos) and post it, labelled as AI-generated.

One photo or two

On the table: your photo (or your pet's) goes on the table and the AI adds a stranger as the guest who walks in.
Both of us: two photos, the first on the table and the second walking in. Two separate photos are fine; you don't need one of you together.

What you get

A 10 or 15-second vertical MP4 up to 720p, with an upbeat house beat over cafe sounds, or no sound (add the trend's own song in TikTok or Instagram when you post), no watermark.
You sit on the tiny table like a stool
A guest walks in and sits the right way
You notice and hop down
You sit properly as if nothing happened

Photos that don't work well

Group photos: one person or one pet per photo, or the AI may pick the wrong face.
Sunglasses, a hat over the eyes or a heavy beauty filter.
A tiny face far away in the frame; waist up or head to toe, face to the camera, works best.

Post it like the trend

The joke is the slow realization, so caption it from your side: POV, you have never been anywhere this fancy.

POV: Your first time in a fancy cafe
Nobody told me the stool was a table
Me pretending I go to places like this

If something goes wrong

If a video fails to generate, its credits go back to your balance automatically and you can try again. If you're not happy with your first paid video, you get a full refund within 7 days. You can delete your photos and videos from My videos at any time.