ddetailer
ddetailer | sd_dreambooth_extension | |
---|---|---|
12 | 115 | |
335 | 1,824 | |
- | - | |
3.8 | 8.7 | |
5 months ago | about 1 month ago | |
Python | Python | |
- | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
ddetailer
-
Can't use controlnet anymore
Looks like it's a known issue. Looks like you'll need to remove the mmcv folder in your venv folder.
-
After training 50+ LoRA Models here is what I learned (TIPS)
One way to avoid the inpainting issue is to use Ddetailer or Adetailer to auto-inpaint detected faces. I use the latter as I couldn't get ddetailer to work, but the fundamental feature is the same. It does a face-detection and inpainting pass after the initial generation (and hires fix), which means your first output already has the face fixed somewhat. And then you can go in to correct smaller details.
- Here are some more hires images I made (max size 4096 * 2752) by webUI
- Restore faces: when?
-
Getting strange error when booting up Automatic1111 but it stills run fine
You can downgrade the mmdet https://github.com/dustysys/ddetailer/issues/41 or move to active fork https://github.com/Bing-su/dddetailer
-
Batch face swap script, see comment
Check also GitHub - dustysys/ddetailer It's not obvious how to adjust it's settings but it's specifically for detecting/segmentig persons and faces
-
Theory about bad small faces. Could it just be how the noise steps are scheduled?
use ddetailer my dude https://github.com/dustysys/ddetailer seems like is exactly what you need
-
1.5 keeps spitting out garbled garbage, Ive tried every single combo of settings, ive tried prompts from my 1.4 logs that gave me good results in the past and ive tried multiple models. Whats going on?
Get DDetailer Makes faces super HD without having to go into "High Resolution Fix", like... everyone needs this, install it on extensions page by pasting the GitHub link.
- the ddetailer extension works amazing with inpainting model!
-
so i am a college student and a total newbie in this, would like some advice :-D
I'm not sure if there is something like this, because you need to manually highlight the things you need to inpaint, but there was extension for auto-masking... Oh, it actually made for inpainting. https://github.com/dustysys/ddetailer or you can go to available tab in extensions.
sd_dreambooth_extension
- SDXL Training for Auto1111 is now Working on a 24GB Card
-
(Requesting Help)
I am trying to use StableDiffusion via AUTOMATIC1111 with the Dreambooth extension
-
it will be an absolute madness when sdxl becomes standard model and we start getting other models from it
When I first attempted SD training, I was very frustrated. It wasn't until I found this obscure forum thread on Github that I actually started producing great results with Dreambooth. Because I have such satisfactory results, I'm very reluctant to beat my brains against LoRa and its related training techniques. I gave up trying to train TI embeddings a long time ago. And I never figured out how to train or how to use hypernetworks. I've only been able to get good results with Dreambooth directly because of that thread I linked above. I make LoRas by extracting them from Dreambooth-trained checkpoints. And I have no idea if I'm doing the extractions the right way or not.
-
"Exception training model: ' Some tensors share memory" with Dreambooth on Vladmatic
Getting the same with automatic1111 and sd_dreambooth extension. Check out more here in the issues log: https://github.com/d8ahazard/sd_dreambooth_extension/issues/1266
-
Yo, DreamBooth gatekeepers, SHARE YOUR HYPERPARAMETERS, please.
It's several moths old and many things have changed. But the spreadsheet available through this thread on Github has been indispensable for me when I train Dreambooth models. I'm astounded no one talks about it. I bring it up all the time. The research presented there should be continued. I'd love to see similar research done for SD v2.1.
-
What is the BEST solution for hyper realistic person training?
Training rate is paramount. Read this Github thread.
-
How do you train your LoRAs, 1 Epoch or >1 Epoch (same # of steps)?
https://github.com/d8ahazard/sd_dreambooth_extension/discussions/547/ (in depth training principles understanding)
-
Struggling to install Dreambooth
sd_dreambooth_extension https://github.com/d8ahazard/sd_dreambooth_extension.git main 926ae204 Fri Mar 31 15:12:45 2023 unknown
- Attempting to train a lora with RTX 2060 6 GB vRAM, how to go about this?
-
SD just released an open source version of their GUI called StableStudio
also the Dreambooth extension supports API (https://github.com/d8ahazard/sd_dreambooth_extension/blob/main/scripts/api.py) so i'm not sure where do you get those news :/
What are some alternatives?
adetailer - Auto detecting, masking and inpainting with detection model.
lora - Using Low-rank adaptation to quickly fine-tune diffusion models.
dddetailer - Detection Detailer hijack edition
kohya_ss
stable-diffusion-webui - Stable Diffusion web UI
kohya-trainer - Adapted from https://note.com/kohya_ss/n/nbf7ce8d80f29 for easier cloning
sd-webui-segment-anything - Segment Anything for Stable Diffusion WebUI
stable-diffusion-webui-wd14-tagger - Labeling extension for Automatic1111's Web UI
civitai - A repository of models, textual inversions, and more
dreambooth-training-guide
stable-diffusion-webui-anti-burn - Extension for AUTOMATIC1111/stable-diffusion-webui for smoothing generated images by skipping a few very last steps and averaging together some images before them.
sd-scripts