This App inserts a face in a target image using a LORA (Flux-based) of your choice that must be capable of drawing a particular face of interest (for example a Marilyn Monroe LORA). Additional mild changes may be done to head orientation, mouth and eyes. Three images (attempts/variations) will be generated.

LORA Strength : Each LORA has its own optimal strength, usually between 0.8 and 1.0. Suggested value is 0.9.

Face selector : If there's more than one face in the target image, choose the one to modify with a number. Faces are enumerated from biggest to smallest starting at zero. If faces are too close together, set the "Tight" option to "Yes" to try avoid distortions.

Face expression : Optional. A prompt to midly adjust the face expression or facial features like eye color. This is largely dependent on the chosen LORA's ability to produce the desired prompt. Write "none" to ignore this option.

Aggresive : The third image generated is always more aggresive in seeking changes, which may lead to distorted results with some LORAs. If you're consistently getting a bad third option, set this option to zero.

Result resolution : Optional. The amount of megapixels wanted in the resulting image, usually between 0.3 and 2. Leave at zero to preserve the original image size.

Mouth options : Optional. Better ignore if the face in the target image already has the right mouth form. They are:
0 = unchanged
1 = wide smile
2 = open mouth
3 = serious

Eyes options : Optional. Better ignore if the face in the target image already has the right eyes form. They are:
0 = unchanged
1 = looking down
2 = looking up
3 = looking left
4 = looking right
5 = looking at viewer

Modifiers for eyebrows, eyelids and head rotations are optional. Leave at zero to ignore them.

Color drawing