Yolov26x segmentation model trained on images of penises. This model was only designed to detect realistic images so it won't work for 2d or anime penises.You can find the 2D version here <a target="_blank" rel="ugc" href="https://civarchive.com/models/310687">https://civarchive.com/models/310687</a><h2 id="trained-on-ultralytics-8.4.47">Trained on ultralytics 8.4.47</h2>You will need to update to that version or newer. If you get the error “Can't get attribute 'Segment26'” Please update.You can update by running ./python -m pip install --upgrade ultralytics&gt;=8.4.47 in your VENV<h2 id="how-to-use-in-comfyui">How to use in ComfyUI</h2><ul><li>Install Impact Pack and Subpack<ul><li>Install ComfyUI-Manager from <a target="_blank" rel="ugc" href="https://github.com/ltdrdata/ComfyUI-Manager">https://github.com/ltdrdata/ComfyUI-Manager</a></li><li>Open the Manger window</li><li>Open Install Custom Nodes</li><li>Search for and install ComfyUI Impact Pack</li><li>Search for and install ComfyUI Impact Subpack</li></ul></li><li>Save the model to ~\ComfyUI\models\ultralytics\segm\</li></ul><h2 id="node-setup">Node setup</h2><ul><li>Add a FaceDetailer node from the Impact Pack</li><li>Add a UltralyticsDetectorProvider node and Select the ADetailer model, connect BBOX_DETECTOR to bbox_detector, and SEGM_DETECTOR to segm_detector_opt</li></ul>

Cock and Ball Detection (ADetailer)

Updated to yolo26x Trained on 3,776 images

Files with SHA256: d3eb09115c7c5a2397981d2a03c1424ddbbab047d1e116aee3ee7220d0dc586d