Felldude/Yolo11_NSFW_Nano

Full Custom YOLO Detection Model

Model Description

This model is a custom-trained YOLO object detection model for multi-class detection and segmentation tasks on a specialized dataset.

Intended Use

• Object detection on NSFW datasets
• Bounding box classification for custom classes

Limitations

• Trained on a custom dataset that was boxed with the ten classes by hand.
• Performance may degrade on unseen domains or distributions, but in testing on 10k out of database images the error rate was less the 4%

Evaluation Results

Overall Metrics

• Precision: 0.795
• Recall: 0.814
• mAP@50: 0.861
• mAP@50-95: 0.530

Classes

breast, vulva, butt, penis, anal, vaginal, blowjob, handjob

Per-Class Results

Class	Images	Instances	Precision	Recall	mAP50	mAP50-95
breast	286	298	0.826	0.830	0.887	0.557
vulva	165	166	0.721	0.693	0.753	0.362
butt	127	131	0.682	0.801	0.792	0.467
penis	237	269	0.719	0.851	0.869	0.559
anal	145	147	0.952	0.905	0.941	0.551
vaginal	182	184	0.850	0.891	0.912	0.572
blowjob	36	36	0.868	0.778	0.907	0.640
handjob	73	88	0.832	0.864	0.919	0.618