Stable Diffusion 1.5 model trained in Dreamboot with 50 images from the web of women's bare feet / soles with 150 steps per image. Initial test results are near perfect (the number of toes and sizes are usually correct!), sometimes left and right get swapped, sometimes arms have feet attached, and some weird results sometimes when trying to specify a character hair color being applied to the feet.
Include 'wbft' in prompt.
Use at your own risk. I am not an author. I do not own the rights to the images used to train this dataset. This is for educational purposes only.