Replace a person
Open Inpaint ,add pic and draw a mask
add Lora to prompt
Generate with settings
CFG scale: body 6-7, after face- 1.5( with lora F), only face-1.5
Denoising strength: 0.6-0.75
Process the edges of the masks and possible errors with CFG scale 1.5, Denoising 0.6
These images were generated with the this prompt on the latest Realistic Vision model
Black and white, retro photo, (noise), ,photography, detailed skin,
realistic, photo-realistic, 8k, full length frame, High detail RAW art,
diffused soft lighting, shallow depth of field, sharp focus,
hyperrealism, cinematic lighting, woman, side picture , <lora:YOUR LORA:0.8>
Negative prompt: <hypernet:bad-hands-5:1>, (((golden ratio))), bw,
(tan skin:1.3),(worst quality:1.5), (low quality:1.5), low-res,
(nose:1.5), (((chromatic aberration))),
<hypernet:realisticvision-negative-embedding:1>,
<hypernet:CyberRealistic_Negative-neg:1>,((blur censor)),
((blurry)), (blurry background), (blurry foreground), bokeh, (chromatic
aberration), closed mouth, cosplay photo, eyelashes, , motion blur,
nose, overexposed, (deformed iris, deformed pupils, semi-realistic, cgi,
3d, render, sketch, cartoon, drawing, anime:1.4), text, close up,
cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly,
duplicate, morbid, mutilated, extra fingers, mutated hands, poorly
drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated,
bad anatomy, bad proportions, extra limbs, cloned face, disfigured,
gross proportions, malformed limbs, missing arms, missing legs, extra
arms, extra legs, fused fingers, 4 fingers, 3 fingers, too many fingers,
long neck
realistic, photo-realistic, 8k, full length frame, High detail RAW art,
diffused soft lighting, shallow depth of field, sharp focus,
hyperrealism, cinematic lighting, woman, side picture , <lora:YOUR LORA:0.8>
Negative prompt: <hypernet:bad-hands-5:1>, (((golden ratio))), bw,
(tan skin:1.3),(worst quality:1.5), (low quality:1.5), low-res,
(nose:1.5), (((chromatic aberration))),
<hypernet:realisticvision-negative-embedding:1>,
<hypernet:CyberRealistic_Negative-neg:1>,((blur censor)),
((blurry)), (blurry background), (blurry foreground), bokeh, (chromatic
aberration), closed mouth, cosplay photo, eyelashes, , motion blur,
nose, overexposed, (deformed iris, deformed pupils, semi-realistic, cgi,
3d, render, sketch, cartoon, drawing, anime:1.4), text, close up,
cropped, out of frame, worst quality, low quality, jpeg artifacts, ugly,
duplicate, morbid, mutilated, extra fingers, mutated hands, poorly
drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated,
bad anatomy, bad proportions, extra limbs, cloned face, disfigured,
gross proportions, malformed limbs, missing arms, missing legs, extra
arms, extra legs, fused fingers, 4 fingers, 3 fingers, too many fingers,
long neck
Modify the prompt data depending on the image. But the neural network understands well what is shown in the picture and is capable of generating pictures of a suitable style and color.
