Meet Marsey! An adorable cat from a Telegram sticker pack. I've been trying to get SD to generate more of this character, and wanted to share my results for anyone else working on a specific 2D style.
Comparisons
a photo of a spaceman Marsey in outer space
Textual Inversion / DreamBooth
a photo of Marsey as a lifeguard
Textual Inversion / DreamBooth
a photo of Marsey as a scientist
Textual Inversion / DreamBooth
a photo of Marsey as a gardener
Textual Inversion / DreamBooth
What I've noticed:
Textual inversion:
Excels at style transfer. "elephant in the style of Marsey"
May benefit from more images. My run with 74 images performed better than the one with 3
Best results (both in terms of style transfer and character preservation) at
25,000 steps
DreamBooth:
Far, far better for my use case. The character is more editable and the composition improves. It doesn't match the art style quite as well, though.
3 images worked better than 72
works extremely well with cross-attention prompt2prompt (the "img2img alternative test" script in automatic1111's UI)
1,000 steps (30min on an A6000) is sufficient for good results
Worth mentioning - it's usable with deforum for animations
Combining the two doesn't seem to work, unfortunately. The next step might be either to directly finetune the network itself and apply one of these techniques afterwards, or possibly training the classifier.
Jump in the discussion.
No email address required.
What would happen if you fed a couple thousand renders of a 3d marsey into textual inversion? 3D renders seems to give good results
@HeyMoon
Jump in the discussion.
No email address required.
I tried this with 5 renderings of the 3D Marsey earlier, it didn't really get it. Maybe bumping the number would help
Jump in the discussion.
No email address required.
Renders would have to be from every angle and have a variety of real backgrounds to avoid overfitting. Just five and it doesn't know how big marsey is supposed to be.
Jump in the discussion.
No email address required.
More options
Context
More options
Context
More options
Context