nexwayne/fursuit-yolo11m

Fursuit Head Detection (YOLO11m)

Custom object detection model based on YOLO11m trained specifically for detecting fursuit heads in images and video streams.

The dataset includes negative background samples (real animals, birds, etc.) to minimize false positives.

---

Model Performance

Metric	Value
mAP@50	0.918 (91.8%)
mAP@50-95	0.656 (65.6%)
Precision	0.940 (94.0%)
Recall	0.845 (84.5%)
Optimal Conf Threshold	0.49

---

Usage Examples

Option 1: Direct Download via Python

from ultralytics import YOLO

# Load weights directly from Hugging Face Hub
model = YOLO("https://huggingface.co/nexwayne/fursuit-yolo11m/resolve/main/best.pt")

# Perform inference with optimal confidence threshold
results = model.predict(
    source="path/to/image.jpg",
    conf=0.49,
    save=True
)

# Show results
results[0].show()