Claquasse/foot_anime_yolo
A small YOLO11m detector that finds feet in anime and illustration images. It has a single class, foot. The main use is image generation cleanup: pair it with an ADetailer or Impact Pack workflow in ComfyUI so a detail or inpaint pass can fix feet, which diffusion models often render badly. It was built for the Anima text-to-image model, but it works on anime-style imagery in general, so it should
pip install mlforge-sdk && mlforge pull Claquasse/foot_anime_yolo
Model details
About Claquasse/foot_anime_yolo
A small YOLO11m detector that finds feet in anime and illustration images. It has a single class, foot. The main use is image generation cleanup: pair it with an ADetailer or Impact Pack workflow in ComfyUI so a detail or inpaint pass can fix feet, which diffusion models often render badly. It was built for the Anima text-to-image model, but it works on anime-style imagery in general, so it should transfer to other anime or illustration generators without retraining.