search

JoePenna / Dreambooth-Stable-Diffusion

Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) by way of Textual Inversion (https://arxiv.org/abs/2208.01618) for Stable Diffusion (https://arxiv.org/abs/2112.10752). Tweaks focused on training faces, objects, and styles.
MIT License
3.2k stars 554 forks source link

Feature/wiring up flip param #149

Closed djbielejeski closed 1 year ago

djbielejeski commented 1 year ago

Adding the ability to turn off flip. This is useful for training a human face.

See https://github.com/victorchall/EveryDream2trainer/blob/main/doc/ATWEAKING.md#flip_p

If you wish for your training images to be randomly flipped horizontally, use this to flip the images 50% of the time:

--flip_p 0.5 ^ This is useful for styles or other training that is not asymmetrical. It is not suggested for training specific human faces as it may wash out facial features as real people typically have at least some asymmetric facial features. It may also cause problems if you are training fictional characters with asymmetrical outfits, such as washing out the asymmetries in the outfit. It is also not suggested if any of your captions included directions like "left" or "right". Default is 0.0 (no flipping)

djbielejeski commented 1 year ago

Just a note, we left flip at 0.5 to not break existing workflows, but added an option to turn it off in the notebooks.