Should I use a selfie or a full-body photo?
A clear face is the first priority for identity. A head-and-shoulders or waist-up image with visible eyes is a practical starting point. A full-body image can provide more clothing context, but a tiny face in a distant shot is less useful.
The generated scene includes sitting and standing. A portrait cannot guarantee accurate legs, hands or unseen clothing. If the source photo does not show an outfit clearly, the model must infer some details.
What should I avoid?
Choose an original image rather than an enlarged, blurry screenshot. Avoid sunglasses, masks, faces cut off at the edge, dramatic colored lighting and beauty filters that substantially change facial features.
- One person per upload, not a crowded group photo.
- No celebrity impersonation, minors or nonconsensual likeness use.
- Do not upload a pet to this adult-portrait template; dog pictures in the studio are role placeholders, not cafe results.
- Re-saving a small image at a larger size does not restore missing facial detail.
What happens when a photo is rejected?
Check the file format and size first. Then choose a clearer photo with one adult face and wait for the new check to finish. Automated screening can make mistakes; repeatedly submitting the same rejected file is unlikely to help.
Only upload images you have permission to use. Photo checks do not prove ownership or consent, and approval is not a guarantee that a generated face will be an exact match.
Explore the protagonist and optional cafe regular roles. Choose your characters, check the final price and create your own wrong-seat moment.
Try the video studio →