Building a photo-to-video window scene without a prompt box

李婧璇 — Fri, 30 Oct 2026 04:19:27 +0000

I have been building a small photo-to-video product called Burger Window AI. The whole product is one scene: upload a bright, front-facing photo of a person or a pet, and it returns a vertical clip of that subject at a night window, mid-bite, staring at the camera, then the curtain closes.

Product: https://burgerwindowai.com

Why one scene, not a prompt box

General video models ask the user to write a prompt. Most of the people this is for are posting to TikTok, Reels, or Shorts. They already have the photo. They do not want to describe lighting, camera, or timing.

So the generator is a fixed scene, not an open prompt:

Warm night window.
A short pause mid-bite.
A look straight into the camera.
Curtain close.
About 12 seconds, vertical MP4.

No prompt writing. The input is the photo.

What the photo has to be

The pipeline is picky about the source frame:

Front-facing
Bright enough that the face or the pet is not in shadow
One clear subject

If the photo is a side profile or a dark room shot, the window scene falls apart because the model has to invent a face it never saw.

What the output is for

The clip is sized for short-form feeds. The beat is simple on purpose: the freeze mid-bite is the hook, the stare is the hold, the curtain is the ending. That is the whole edit. People can post it as-is.

What I am not doing

This is not a general text-to-video studio. One scene, one upload, one MP4. The tradeoff is obvious: you cannot ask it for a different location or a different action. If you want the window scene, that constraint is the product.

If you are building something similar, a narrow scene with a strict photo spec is easier to ship than a prompt box that can do anything and usually does it wrong.

https://burgerwindowai.com

Building a photo-to-video window scene product

李婧璇 — Fri, 30 Oct 2026 03:33:31 +0000

I have been building Burger Window AI, a small product that turns one front-facing photo of a person or a pet into a vertical, roughly 12-second night-window clip.

The scene is fixed on purpose. There is no prompt box. You upload a bright photo, and the clip plays the same beat every time:

A warm night window.
A mid-bite freeze.
A stare into the camera.
The curtain closes.

That constraint is the product. Short-form feeds (TikTok, Reels, Shorts) reward a recognizable format more than a general video model. A fixed scene also keeps the pipeline simple: one photo in, one MP4 out, about 12 seconds, vertical.

A few things that shaped the build:

A single camera angle and a short duration are easier to keep consistent than an open-ended generator.
The photo has to be front-facing and well lit, or the face and the bite do not read in the first second.
Shipping a scene, not a prompt UI, cuts the instructions down to "upload a photo."

The live product is at https://burgerwindowai.com