Since nobody else seemed to have made one...I present my Illustrious character LoRA for Dean McCoppin from The Iron Giant.
V2 is my second version and produces good results. The previous version had AI images made using Pony so I removed those, remade the NSFW images using Illustrious, and kept AI images to a minimum so the training would be based more on the movie images. I tried changing the facial hair tag but soul patch does NOTHING in stable diffusion (even though it exists in the general booru tag list) so I had to stick with goatee.
If you want the generations to more closely resemble the movie version just put realistic in your negative prompt. You can also add cel-shaded to change things up but that tag was not in my training.
Recommended LoRA strength: 0.7
Also, I use an adetailer in my generations for this LoRA.
V2 main tags are as follows:
dmccoppin, 1boy, black hair, short hair, brown eyes, goateeV2 outfit tags are as follows:
black turtleneck, sleeves rolled up, black belt, jeans, black shoesblack leather jacket, black turtleneck, black belt, jeans, black shoesV1 optional tags:
sunglassesThe settings I used for my generation testing were 4-5 CFG, 20 steps, Euler - Karras on Forge with an adetailer (face_yolov8s).
V1 is my initial version and produces great results. I only did his main outfit since it was hard enough to get decent screenshots directly from the movie.
Recommended LoRA strength: 0.7
Also, I use an adetailer in my generations for this LoRA.
V1 main tags are as follows:
dmccoppin, 1boy, black hair, short hair, brown eyes, goateeoptional - add "muscular" to negative prompt if he looks too bulky (especially in NSFW or shirtless shots).
V1 outfit tags are as follows:
black turtleneck, sleeves rolled up, black belt, jeans, black shoesblack jacket, black turtleneck, black belt, jeans, black shoesV1 optional tags:
sunglassesThe settings I used for my generation testing were 3-5 CFG, 20 steps, DPM++ 2M SDE - Karras on Forge with an adetailer (face_yolov8s).
Description
V2 is my second version and produces good results. The previous version had AI images made using Pony so I removed those, remade the NSFW images using Illustrious, and kept AI images to a minimum so the training would be based more on the movie images. I tried changing the facial hair tag but soul patch does NOTHING in stable diffusion (even though it exists in the general booru tag list) so I had to stick with goatee.
If you want the generations to more closely resemble the movie version just put realistic in your negative prompt. You can also add cel-shaded to change things up but that tag was not in my training.