All credit to sevenof9247
ATTENTION -> weight correction 0.3-0.9 results with 25 steps DPM++2M SDE Karras almost the same images as without and 70 steps
Trigger is not essential but useful for the culture ! ;)
has a great impact on lower Sampling steps <30
if you use DPM++2M SDE Karras with 70 steps, you meight dont need my lora
v2 -> much more smother and better ;)
please merge in your checkpoint, and tell me, btw i have less trained loras
more hints below! ->
face enhancer
please rate it if you like for the contest (thx) and show me your results...
(learning: ran on an intel-i7 with an RTX4060, 7 hours with 300 images)
its a lora file, so copy in your lora folder and use it like usual
often it works without changing the face too much, i suggest weights 0.2 until 0.5, the higher the less background and it works more for closer faces!
sometimes minor improvements are possible through sampling steps 30 up to 60
-> for faces in more distance i suggest ADetailer (face) without an additional prompt
SDXL new VAE (2023.07.26) is quite better than older ones for faces,
but try my lora and you will see often more real faces, not that blurred soft ones ;)
I tried to include many cultures, 11-if i remeber^^ with old and young content, at the moment only woman
Hints:
brackets and numbers do not work like in SD1.5 !
"professional photo" is similar to "realistic"
"masterpiece", ok for a promt with only 3-4 words it changes the image a lot, if you have more than 8-9 words the image turn in to warm tones and may more details in background
All SFW!
