A model trained to do the superhero domino mask. Tends to work best at around 0.6. This model tends to make a mask that matches the outfit but defaults to a normal black one. Doesn't mix great with other LORA's but if trying to do so consider adding some modifiers to the prompt directly why playing with it (Ex: (domino mask:1.2)). This model was trained on WarriorMama777's orange mix from huggingface. Made with the Kohya tool for AI Training. Any suggestions to improve or recommended things for the data set would be appreciated.
After several attempts this is the first version that doesn't distort the face. and this version tends to prefer a thin mask that goes around the eyes rather than one that covers the entire forehead. This actually was a quite a bit more difficult than expected so any suggestions or just good ideas for future data sets would be appreciated, enjoy.
Go ahead and upload yours!
Your query returned no results β please try removing some filters or trying a different term.