Not-For-All-Audiences
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,15 +1,18 @@
|
|
| 1 |
# NSFW Segmentation
|
| 2 |
|
| 3 |
-
Multi-head release of
|
| 4 |
|
| 5 |
-
| File | Task | Classes | Mask mAP@0.5 | Mask mAP@0.5:0.95 |
|
| 6 |
-
| --- | --- | --- | --- | --- |
|
| 7 |
-
| `nsfw-seg-breast.pt` | Breast anatomy | breast, areola, nipple | 0.895 | 0.636 |
|
| 8 |
-
| `nsfw-seg-
|
| 9 |
-
| `nsfw-seg-
|
|
|
|
|
|
|
|
|
|
| 10 |
|
| 11 |
## Description
|
| 12 |
-
-
|
| 13 |
- Weights exported as `.pt` checkpoints compatible with `ultralytics>=8.3`.
|
| 14 |
- One model per label space; load the checkpoint that matches your target anatomy.
|
| 15 |
|
|
@@ -36,7 +39,7 @@ Multi-head release of three single-task segmentation models targeting NSFW anato
|
|
| 36 |
```python
|
| 37 |
from ultralytics import YOLO
|
| 38 |
|
| 39 |
-
model = YOLO("nsfw-seg-breast.pt") # swap for
|
| 40 |
results = model.predict("path/to/image.jpg", imgsz=832, conf=0.1)
|
| 41 |
```
|
| 42 |
|
|
|
|
| 1 |
# NSFW Segmentation
|
| 2 |
|
| 3 |
+
Multi-head release of single-task segmentation models targeting NSFW anatomy. Each checkpoint runs independently and produces binary masks for the specified classes.
|
| 4 |
|
| 5 |
+
| File | Backbone | Task | Classes | Mask mAP@0.5 | Mask mAP@0.5:0.95 |
|
| 6 |
+
| --- | --- | --- | --- | --- | --- |
|
| 7 |
+
| `nsfw-seg-breast-s.pt` | YOLO11s | Breast anatomy | breast, areola, nipple | 0.895 | 0.636 |
|
| 8 |
+
| `nsfw-seg-breast-x.pt` | YOLO11x | Breast anatomy | breast, areola, nipple | 0.929 | 0.702 |
|
| 9 |
+
| `nsfw-seg-vagina-s.pt` | YOLO11s | Vagina | vagina | 0.995 | 0.871 |
|
| 10 |
+
| `nsfw-seg-vagina-x.pt` | YOLO11x | Vagina | vagina | 0.995 | 0.918 |
|
| 11 |
+
| `nsfw-seg-penis-s.pt` | YOLO11s | Penis | penis | 0.995 | 0.975 |
|
| 12 |
+
| `nsfw-seg-penis-x.pt` | YOLO11x | Penis | penis | 0.995 | 0.987 |
|
| 13 |
|
| 14 |
## Description
|
| 15 |
+
- Backbones: YOLO11s and YOLO11x segmentation heads (Ultralytics 8.3.204).
|
| 16 |
- Weights exported as `.pt` checkpoints compatible with `ultralytics>=8.3`.
|
| 17 |
- One model per label space; load the checkpoint that matches your target anatomy.
|
| 18 |
|
|
|
|
| 39 |
```python
|
| 40 |
from ultralytics import YOLO
|
| 41 |
|
| 42 |
+
model = YOLO("nsfw-seg-breast-s.pt") # swap for -x or other anatomy
|
| 43 |
results = model.predict("path/to/image.jpg", imgsz=832, conf=0.1)
|
| 44 |
```
|
| 45 |
|