Actually, this model is not good enough in the form of LoRA; it should be presented in the form of a controlnet model. However, my device limits my training. If you're interested, you can try this dataset (Anime) :

https://huggingface.co/datasets/Makki2104/difference_images_Cloth-Nude

This model is trained based on the Illustrious-XLv0.1 model. Please use it with the Illustrious-XLv0.1 model and its derivative versions(recommend:NoobAI-XL (NAI-XL)).

The model is designed to remove clothing from characters while keeping other parts mostly unchanged. A picture is worth a thousand words—please refer to the preview image for more details.

Test environment: You can obtain the ComfyUI workflow through the image below.

Warning:

This is currently a beta version, and in some cases, it may not function properly. I am working on training with a larger dataset, and the training method and dataset will be made public in the future. Stay tuned!

2024/12/20:

update gamma version

Upload datasets:

https://huggingface.co/datasets/Makki2104/difference_images_Cloth-Nude

train method:

https://makkishizu.github.io/2024/12/13/%E5%A6%82%E4%BD%95%E8%AE%AD%E7%BB%83%E5%B7%AE%E5%BC%82%E4%B8%B9/