trigger word : emwt
recommended checkpoint : Realistic Vision 4 No VAE
Training data included : Images used + Captions
if you like my stuff pls consider supporting me on Kofi
accepting private Commissions on my disc : dogucat as well...
Show More
Description
FAQ
Comments (9)
Would like to see 2010 pixie cut emma specifically.
If you write in your prompt ....
" A full Body Portrait Photo of a Emma Watson with a Pixie Hair Style, .... "
Your A.I. will surely create for you a Pixie hair cut style.
If you write a color for it ,example " with a Red Pixie Hair Style"
Your A.I. will also color it
@maxhitman yeah, prompting is very imp
@dogu_cat That is correct.
" A full Body Portrait Photo of a actress Emma Watson with a Dirty Blonde Pixie Hair Style, with a long moustache and fuzzy beard, Beautiful, perfect eyes, (high detailed skin:1.1), perfect SLIM body, wearing a black Latex BDSM bodysuit , Modelshoot pose, Professional Photography, soft lighting, Photorealistic, Realistic, Standing in a Dungeon room inside a creepy Castle background, blurred background, RAW, analog style, Erotic, sharp focus, 8k, high res, DSLR, high quality, Fujifilm XT3, film grain, award winning, masterpiece "
LMAO
@maxhitman flagged for review recipe XD
May I make an important suggestion?
Since we already know that this SD A.I. can create very good images of Emma W. without the use of any Lora, Lycoris or T.I.,
I also agree that some times these pre-trained models can be useful when using Checkpoints which need help to create her face.
But from all the models already posted here, I am not able to see one model of Emma´s face in which she looks like in 2022 or 2023.
She has grown up and is a very beautiful woman now. Her face looks very different now from the time she was in the "Harry Potter" series.
Cheers
MAX
Very true Indeed, I sorta gave it a blend with her in Deathly hallow part 1 where she was 18/19, and some photoshoots of her in 2015/16,
So her Harry Potter look + a dash of her later change,
I will try for her 2023 look in next update
@dogu_cat Yes, I have noticed you trained the model with some Mixed images.
It is suggested by many model-creators to train a model using images from a certain Age , or Year you want the person to be, so the training comes out perfect and the model comes out perfect.
This lora does the job, but only with really small prompts like what op has here. Trying to get more complex with facial expressions or different settings makes it start to lose realism. Works well, but isn't flexible.
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.