<h1 id="breastbox-detection-(yolov8s+)">🎯 BreastBox Detection (YOLOv8s+)</h1>The Precision Tool for Breast Detection in Stable DiffusionBreastBox is a custom-trained object detection model (YOLOv8s+) designed specifically to solve common detection failures in ADetailer. While default models often struggle with side views, overlapping clothing, or dynamic poses, BreastBox provides a tighter, more anatomically correct bounding box.<hr /><h3 id="why-use-breastbox">🔥 Why Use BreastBox?</h3><ul><li>✅ High Precision: Manually curated dataset, moving from 200 to over 650 high-quality labels.</li><li>✅ Challenging Angles: Expertly handles "lying down", "side-boob", "from behind", and partially obscured views.</li><li>✅ Anatomically Correct: Designed for tighter masks that follow the natural shape of the body.</li><li>✅ Style Agnostic: Optimized for 3DCG (DAZ/Poser), but generalizes well to realistic and 2D/Anime styles.</li></ul><hr /><h3 id="version-history">📊 Version History</h3><ul><li>v1.0 (Experimental): Initial release (~200 samples). Best for standard poses.</li><li>v2.0: Major data upgrade (~400 samples). Better versatility for breast sizes and camera angles.</li><li>v3.0+ (Latest): Focus on obscured breasts (hands, objects) and smaller chest sizes (~700+ samples).</li></ul><hr /><h3 id="quick-start-(adetailer)">🚀 Quick Start (ADetailer)</h3><ol><li>Download the<pre><code>.pt</code></pre>file.</li></ol><ol><li>Move it to:<pre><code>stable-diffusion-webui\models\ADetailer</code></pre></li><li>Select<pre><code>BreastBox</code></pre>from the ADetailer dropdown in WebUI.</li></ol><ol><li>Recommended Settings:<ul><li>Detection Confidence:<pre><code>0.3</code></pre>to<pre><code>0.5</code></pre></li></ul><ul><li>Mask Merge Mode:<pre><code>Merge</code></pre></li></ul></li></ol><hr /><h3 id="the-power-behind-the-training:-yolo-training-gui">🛠️ The Power Behind the Training: YOLO Training GUI</h3>I trained and refined this model using YOLO Training GUI, a standalone tool I developed to make YOLO training visual and simple for everyone.If you are training your own models and want to see your results like I do, check it out: 👉 <a target="_blank" rel="ugc" href="https://github.com/Hasher95/YOLO-GUI">YOLO Training GUI on GitHub</a>Why you'll love it:<ul><li>Live Monitor: Watch detections happen in real-time as the model trains.</li><li>Side-by-Side Comparison: Generates the dual-view comparison images.</li><li>No Code Required: A full GUI to manage your datasets, hyperparameters, and exports.</li></ul><hr /><h3 id="feedback-and-community">💬 Feedback &amp; Community</h3>This is a passion project! If you encounter edge cases or have suggestions:<ul><li>Share your Feedback in the comments.</li><li>Post your Example Images (it helps me refine the dataset!).</li><li>If you like the model or the training tool, a Star on GitHub or a review here is much appreciated! ⭐</li></ul>

BreastBox - ADetailer breast detection

<ul><li>Added more data for the training (700+ samples)</li><li>Base model trained Yolov11x</li><li>~ 300 epochs in grid detection + ~ 200 epochs in normal mode</li></ul>

Files with SHA256: 54041672a9986f30524146847aa9a493489b63c8db3e106b3dc7617e4278e818