IF YOU ARE NEW TO STABLE DIFFUSION OR YOU HAVE PROBLEMS WITH MY MODELS GO HERE: Guide
RIGHT NOW IMAGES METADATA ARE MISSING AS I'M USING COMFYUI TO SIMPLIFY THE IMAGE GENERATION. I'M TRYING TO FIND A WAY TO FIX THIS, IF SOMEONE KNOWS HOW TO DO IT, I'D LIKE TO KNOW.
UPDATE: I've decided to change the name of the LoRA from Renaissance to Painted as the style is too far from the source material I've used for the training. I'll try to make it better in v2.
What's this LoRA
This is a LoRA trained on renaissance paintings depicting women.
The images in the examples are created using only this LoRA and different models (mainly my custom:
You don't need to specify a word to trigger this LoRA.
The LoRA works also on NSFW (check out the examples).
Guides
Learn how to create pictures like mine with my step-by-step tutorials:
If you want to create images with a high LoRA weight, I highly suggest you to check the tutorial.
Support
You can now support me and my work by subscribing to my Deviantart account. You'll also get awesome rewards:
If you want to support my work for free, you can also check out my music/art here:
Description
FAQ
Comments (8)
I love anyone trying to make models, especially ones that aren't another anime model... but these just look like DAZ renders with photoshop filters on them. If you want to make a renaissance style fintetune why not do it with actual renaissance art ? If you want fleshy women, just train it on the work of Ruebens.
Hi! My goal in training my models is not to mimic a specific style, but to create diverse and interesting outputs. I used an anime model as the basis for this LoRA, and I also applied it to anime/cartoon checkpoints for the examples. That's why the results look like a blend of anime and painting.
Important question - Does this create a photo-realistic picture into looking like a Rennaissance artwork? This is the type of thing I am looking for. I like Anime art, but not for this style of artwork.
We REALLY NEED something like this to create other forms of artworks.
BTW... Thank you for your tutorials! I have learned good lessons from them and passed the links to other friends who want to learn.
@maxhitman I used only 8 images of Renaissance paintings from the web to train this LoRA, so it might not capture the style very well. I usually train my LoRAs with small datasets to get close to the original style, but not exactly. Training LoRAs with large datasets takes a lot of time, effort and computing power, which I don't have right now. This is just a hobby for me to have fun and learn new things. I'm also playing with LoCon and LoHa, so maybe I'll try to create something more faithful to the Renaissance style with them.
P.S. I'm happy that you enjoy my articles. I love writing them and I'm glad that they are helpful.
@Inzaniak Thank you once again good friend and continue your Fun adventures in using S.D. !
Cheers
MAX
This doesnt really look renaissance, looks nice tho but I would rename it or something. Looks more vintage style or something.
类似80年代的旧油画风？
Really, really good style, please add more non-toon ones!