A high-precision (92.3%!) small yolov8 model trained to detect all kinds of worn rings.If you want a global jewelry detection model, <a rel="ugc" href="https://civarchive.com/models/1177325/adetailer-jewelry-model">check this out</a>!<h2 id="performance-metrics:-p1mr7rgl4">📊 Performance Metrics:</h2><ul><li>Precision (B): 0.923</li><li>Recall (B): 0.911</li><li>mAP@50: 0.948</li><li>mAP@50-95: 0.661</li></ul><h2 id="usage:-5vsxofvma">ℹ️ Usage:</h2>Recommend using a bbox threshold of 0.50 - 0.60.<h2 id="limitations:-t1tiyzeew">❌ Limitations:</h2><ol><li>Bracelets and watches tend to cause false positives</li><li>Eyes can cause false positives (rare)</li><li>It works with anime style but with less confidence. </li></ol>

Files with SHA256: af127a67691bdf3fa0156f3f3d09712b63bf932d1b3275d0e8eb5c22fdaa28f5