How to choose a photo for AI portraits

A portrait studio is only as good as the photo you give it. The model needs to see your face clearly, understand your proportions, and keep your features consistent across every look. You do not need a professional camera — a clean phone photo is enough.

What makes a good source photo

• One person, clearly visible. A solo portrait gives the model an unambiguous subject.
• Face in focus. Eyes, nose, mouth and jawline should be sharp, not softened by motion blur.
• Even, natural light. Window light or soft indoor light works better than a hard flash.
• A neutral expression and angle. A front-facing or slightly three-quarter view is easiest to carry into different poses.
• Ordinary clothes are fine. The outfit is replaced by the costume you pick; what matters is the face and body shape.

A head-and-shoulders photo is enough for half-body looks. For full-body poses, a photo that shows your proportions works better.

What to avoid

• Group photos, heavy filters, beauty-mode smoothing, or stickers over the face.
• Sunglasses, masks, or hair covering the eyes.
• Extremely dark or backlit photos where the face is a silhouette.
• Photos of someone else. Only upload a photo of yourself, or one you have explicit permission to use.

Before you upload

The studio asks you to confirm that the photo is you or that you are authorized to use it. That confirmation is not a formality: it is the boundary between a creative tool and a misuse of someone else's face.

AI portraits are creative images, not identity documents. The model can change small details, and the result may not match you exactly. Start with drafts, keep the ones that feel like you, and refine only those.

If a result does not look right, try a sharper source photo before changing every setting. The source is the strongest signal you can give the model.