How to make it
Hotel Lobby AI Video: How to Create a Two-Person Performance
Turn two approved reference photos into a short Hotel Lobby AI video with stable identities, positions, camera, and motion.
Updated 2026-10-03
A reliable generation brief
Start with the output requirement, then define the scene. A useful brief says that image 1 is the left performer and image 2 is the right performer, that both positions remain fixed, and that the orange studio, central microphone, lighting, and camera behavior stay stable.
The prompt should then describe a small action loop: the left performer leads, the right performer reacts, and they alternate. This gives the model a sequence without asking it to invent a complex story or change the set.
Source image checklist
- Face visible and reasonably sharp
- One person or character per image
- Minimal sunglasses, masks, or heavy filters
- Similar crop and lighting when possible
- No third-party logos or private details that should not be reproduced
Quality review
Review more than the first frame. Compare each performer at the beginning and end, watch the microphone during hand movement, and listen for audio drift. If the output fails, change one variable at a time so you know whether the source image, prompt, duration, or model caused the problem.
When your references are ready, try one free Hotel Lobby AI Video.
Frequently asked questions
How many reference photos do I need?
A two-person workflow normally starts with one clear reference photo per performer. Use images you own or have permission to use, and label which photo belongs on the left and right.
Why do the two faces merge?
Face merging often comes from ambiguous image roles, similar or low-quality source photos, crowded framing, or too much motion. Use clearer images, explicit LEFT and RIGHT instructions, and a simpler first draft.
Keep exploring
Ready to make your own?
Turn two reference photos into a Hotel Lobby AI video.