Datasets:
license: creativeml-openrail-m
task_categories:
- text-to-image
- keypoint-estimation
language:
- en
tags:
- controlnet
- poses
pretty_name: Duskfall Pose Packs
size_categories:
- 1K<n<10K
Ok SO, as usual we don't usually have time to test these, though there is a chance many of the poses are inclusive of testing - the images are in the grids, and we've included sample images for the face landmarks.
...Don't mock us LOL, we literally found a face landmark demo on huggingface, and went nuts making dumb faces for y'all. "DEPICTS AN ACTUAL PERSON" is because some of the data is AI generated and some of it's realistic people used for poses -- like our face.
Lisc Restrictions: Don't resell any of the sample images in the model card. What you do with the data beyond that is literally not up to use, tho we'd prefer you don't "SELL" the poses. We're not liable for any copyright infringment or any dirty nasty things you make with the poses XD
We're also not liable if easy diffusion tells you that you can merge a single image into 20gb models because you can't, if it tells you that you can then you're gullible LOL.
These packs are copied over from civit as a backup :3 Feel free to use.