arXiv 2504.142212025-04-19cs.CV

Real-IAD D3：面向工业异常检测的真实世界2D/伪3D/3D数据集

Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection

Fudan University（复旦大学）；Shanghai Jiao Tong University（上海交通大学）；YouTu Lab, Tencent（腾讯YouTu实验室）；Rongcheer Co., Ltd.（荣创科技有限公司）；Shanghai Ocean University（上海海洋大学）；Suzhou University（苏州大学）

Wenbing Zhu, Lidong Wang, Ziqing Zhou, Chengjie Wang, Yurui Pan, Ruoyi Zhang, Zhuhao Chen, Linjie Cheng, Bin-Bin Gao, Jiangning Zhang, Zhenye Gan, Yuxie Wang, Y… 展开作者

Wenbing Zhu, Lidong Wang, Ziqing Zhou, Chengjie Wang, Yurui Pan, Ruoyi Zhang, Zhuhao Chen, Linjie Cheng, Bin-Bin Gao, Jiangning Zhang, Zhenye Gan, Yuxie Wang, Yulong Chen, Shuguang Qian, Mingmin Chi, Bo Peng, Lizhuang Ma

备注：: 13 pages. Dataset and code: https://realiad4ad.github.io/Real-IAD D3

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对工业异常检测中多模态数据集稀缺且规模分辨率不足的问题，提出Real-IAD D3高精度多模态数据集，含RGB、微米级3D点云及光度立体伪3D模态，覆盖20类更细缺陷，并设计融合多模态信息的检测方法，显著提升检测鲁棒性与性能。

摘要：

工业异常检测（IAD）日益增长的复杂性已将多模态检测方法定位为机器视觉研究的一个焦点领域。然而，专门为IAD量身定制的多模态数据集仍然有限。像MVTec 3D这样的开创性数据集通过整合RGB+3D数据，为多模态IAD奠定了重要基础，但由于规模和分辨率的限制，在弥合与真实工业环境之间的差距方面仍面临挑战。为了解决这些挑战，我们引入了Real-IAD D3，这是一个高精度多模态数据集，其独特之处在于通过光度立体技术生成额外的伪3D模态，同时包含高分辨率RGB图像和微米级3D点云。Real-IAD D3在20个类别中具有更精细的缺陷、多样化的异常和更大的规模，为多模态IAD提供了一个具有挑战性的基准。此外，我们提出了一种有效的方法，整合RGB、点云和伪3D深度信息，以利用每种模态的互补优势，增强检测性能。我们的实验强调了这些模态在提升检测鲁棒性和整体IAD性能方面的重要性。该数据集和代码可公开访问，供研究使用，网址为https://realiad4ad.github.io/Real-IAD D3。

英文摘要：

The increasing complexity of industrial anomaly detection (IAD) has positioned multimodal detection methods as a focal area of machine vision research. However, dedicated multimodal datasets specifically tailored for IAD remain limited. Pioneering datasets like MVTec 3D have laid essential groundwork in multimodal IAD by incorporating RGB+3D data, but still face challenges in bridging the gap with real industrial environments due to limitations in scale and resolution. To address these challenges, we introduce Real-IAD D3, a high-precision multimodal dataset that uniquely incorporates an additional pseudo3D modality generated through photometric stereo, alongside high-resolution RGB images and micrometer-level 3D point clouds. Real-IAD D3 features finer defects, diverse anomalies, and greater scale across 20 categories, providing a challenging benchmark for multimodal IAD Additionally, we introduce an effective approach that integrates RGB, point cloud, and pseudo-3D depth information to leverage the complementary strengths of each modality, enhancing detection performance. Our experiments highlight the importance of these modalities in boosting detection robustness and overall IAD performance. The dataset and code are publicly accessible for research purposes at https://realiad4ad.github.io/Real-IAD D3

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。