# Medical Skin Anatomy & Clothing Segmenter (YOLOv8-seg)

This is a YOLOv8m-seg (instance segmentation) model trained to detect and segment 15 anatomical regions and clothing coverage states.

It is designed for use in content moderation pipelines, medical imaging research, or with the ADetailer extension in Stable Diffusion (Automatic1111 / ComfyUI) to automatically mask or inpaint specific regions of the body.

---

### ⚠️ Important Note on Class Names

For platform compliance, the class names are published with neutral/medical terminology.

The internal model names are different. If you load the weights .pt or .onnx) directly in Python, you can print the original class names using model.names.

### Classes (15 total):

* 0: body_f

* 1: pelvis_m

* 2: face_f

* 3: pelvis_f

* 4: gluteus_f

* 5: body_m

* 6: chest_f

* 7: lower_pelvis_f

* 8: covered_chest_f

* 9: covered_pelvis_f

* 10: face_m

* 11: covered_gluteus_f

* 12: chest_m

* 13: gluteus_m

* 14: lower_pelvis_m

---

### 🚀 How to use with ADetailer (Stable Diffusion)

1. Download the yolo_v5_best.pt file.

2. Put the file inside your stable-diffusion-webui/models/adetailer/ folder.

3. In the WebUI, enable ADetailer and select yolo_v5_best.pt.

4. Configure your mask settings based on the class indexes above to inpaint or censor specific areas.