Trained a new Stable Diffusion XL (SDXL) Base 1.0 DreamBooth model.
Used my medium quality training images dataset.
The dataset has 15 images of me.
Took pictures myself with my phone, same clothing.
Used the latest Kohya SS GUI and very best DreamBooth SDXL config shared on my Patreon.
Still working great.
Did 150 repeating 1 epoch on my local computer RTX 3090 TI.
Text Encoder is trained too.
Used my very best man regularization images dataset as well.
After Detailer (ADetailer) extension is used.
The images are raw 1024x1024.
The training dataset do not have any full body shot images.
prompt is as below
closeshot photo of ohwx man wearing an expensive {red|green|blue|white|black|yellow|orange|grey|brown|tan|navy} suit in an expensive modern studio, hd, hdr, 2k, 4k, 8k, canon, kodak
What you need to replicate?
Become A Master Of SDXL Training With Kohya SS LoRAs — Combine Power Of Automatic1111 & SDXL LoRAs
How To Do Stable Diffusion XL (SDXL) DreamBooth Training (Full Fine Tuning) On Windows and RunPod
Patreon post for config files ⤵️
https://www.patreon.com/posts/89213064
Reg images datasets⤵️
https://www.patreon.com/posts/massive-4k-woman-87700469
Top comments (0)