Make a Hotel Lobby video with two photos
The Hotel Lobby AI Video Generator turns two reference photos into a short, silent duo performance. It edits a built-in Hotel Lobby reference clip, assigning one uploaded subject to the left performer and the other to the right. This is useful when you want to make a playful video with a friend, partner, family member, pet, or an original cartoon character without filming a scene together.
The editor is available on this page. Add both photos in the studio above, check the left and right roles, and confirm that you have permission to use the images. Signing in opens over the editor. If a later checkout handoff reloads the page, your browser may clear the photo selections; keep the original files available and reselect them if their previews are missing when you return. When a video is ready, preview it here, download the MP4, or use the share option.
Set the left and right performers before creating
The left and right slots define which uploaded reference is mapped to each performer in the source clip. Check the labels and previews before you submit. If the people are reversed, use the swap control to exchange them. Your consent checkbox confirms that you have permission to use both images; only upload photos or artwork you are authorized to use, and get the subject’s permission when appropriate.
The current generator supports person-to-person, person-to-pet, and person-to-original-cartoon pairings. Pet and cartoon results are guided by the appearance in the reference: the tool aims to retain visible identity, species or character design, colors, and accessories. These are generative edits, so details can vary between outputs. A familiar face, outfit, gesture, or frame is not guaranteed to be reproduced exactly.
This is a reference-video edit, not open-ended random generation
A Hotel Lobby generation follows the built-in reference video. The model replaces the two performers using your uploaded references and adapts the result to the template. You are not writing a free-form prompt to invent a different location, script, camera plan, or arbitrary sequence from scratch. That distinction makes the workflow straightforward, while also setting a clear limit on what this page can create.
The system tries to preserve the reference scene and timing while changing the performers. AI video can still alter faces, clothing, proportions, or motion. It may not reproduce a particular dance move or expression precisely, and the output can contain visual artifacts. Review the finished clip before sharing it, especially if the people or characters need to be recognizable.