Institute for AI<p>MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and Segmentation in Zero-Shot Learning</p><p>In manufacturing, quality control remains a critical yet complex task, especially when multiple defect types are involved. MultiADS introduces a system capable of detecting and segmenting a wide range of anomalies (e.g., scratches, bends, holes), even in zero-shot settings.</p><p>By combining visual analysis with descriptive textual input and using a curated Knowledge Base for Anomalies, MultiADS generalizes to unseen defect types without requiring prior visual examples and consistently outperforms state-of-the-art models across several benchmarks, offering a robust and scalable solution for industrial inspection tasks.</p><p>Sadikaj, Y., Zhou, H., Halilaj, L., Schmid, S., Staab, S., & Plant, C. MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and Segmentation in Zero-Shot Learning. International Conference on Computer Vision, ICCV 2025, Hawai, Oct 19-23, 2025, <a href="https://xn--baw-joa.social/tags/ICCV2025" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ICCV2025</span></a>. <a href="https://arxiv.org/abs/2504.06740" rel="nofollow noopener" translate="no" target="_blank"><span class="invisible">https://</span><span class="">arxiv.org/abs/2504.06740</span><span class="invisible"></span></a>.</p><p><a href="https://xn--baw-joa.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AI</span></a> <a href="https://xn--baw-joa.social/tags/AIResearch" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AIResearch</span></a> <a href="https://xn--baw-joa.social/tags/ComputerVision" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ComputerVision</span></a> <a href="https://xn--baw-joa.social/tags/AnomalyDetection" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>AnomalyDetection</span></a> <a href="https://xn--baw-joa.social/tags/ZeroShot" class="mention hashtag" rel="nofollow noopener" target="_blank">#<span>ZeroShot</span></a></p>