per T-29-08 (no `raw`/`html_safe` on YAML). %>

How to Pick a Photo for an AI Portrait

Great portraits start with great source photos. This guide explains the six qualities we look for in a Cardgen upload and how to improve a photo that's almost good enough.

By Cardgen Editorial Team · Last updated 2026-04-09

Before you start

✓ At least one candidate photo
✓ Access to a phone or camera to reshoot if needed

Expected outcome: A photo that will deliver a crisp, recognizable AI portrait on the first try.

Steps

1
Prioritize face sharpness

The face is the most important part of the frame. It should be in focus, with clear eyes and a visible mouth. If the face is motion-blurred or soft, the AI will inherit that softness. Swap to a sharper photo if possible.
2
Check the lighting

Even, natural light is best. Golden hour outdoor photos and window-lit indoor photos both work. Avoid harsh flash with deep shadows under the eyes, and avoid backlit silhouettes where the face is in shadow.
3
Frame the face

The face should occupy roughly 30 to 50 percent of the frame. Too tight and there's no room for styling. Too wide and the face loses detail. Three-quarter body shots and head-and-shoulders portraits both work well.
4
Remove heavy accessories

Sunglasses, face paint, and masks hide the features the AI uses to build the portrait. If possible, pick a photo without them. Glasses are fine — the model handles them well.
5
Upload and preview

Drop the photo into Cardgen and generate a free preview. If the result is soft, regenerate with a different photo from your camera roll — it costs nothing to try a second source.