发表机构
Jeonbuk National University; Korea Institute of Industrial Technology (KITECH); Chung-Ang University（全北国立大学; 韩国生产技术研究院; 中央大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
GRC-Net通过全局注意力MLP和稳定重建模块，在无监督多模态异常检测中实现全局表示一致性，减少正常区域重建误差，并在MVTec 3D-AD和Eyecandies上取得领先性能。
AI 中文摘要
自动化质量检测对于确保产品可靠性至关重要。虽然基于图像的方法能有效捕获外观相关的缺陷，但这些方法在检测结构和几何异常方面存在局限，从而推动了结合3D信息的多模态方法的发展。然而，现有方法主要依赖于局部块级表示，这即使在正常区域也常常导致不稳定的重建误差。为解决这一局限，我们提出了GRC-Net，它集成了一个全局注意力MLP，以在块嵌入间强制实现全局表示一致性，并配备一个稳定重建模块以提高重建稳定性。所提方法通过全局标记捕获整体上下文信息，并通过最小化原始与预测嵌入之间的差异来抑制重建噪声。在MVTec 3D-AD和Eyecandies上的实验表明，GRC-Net在图像级和像素级均持续优于现有方法。定性结果进一步展示了正常区域重建误差的减少以及正常与异常区域之间更明显的重建差异。
英文摘要
Automated quality inspection is essential for ensuring product reliability in manufacturing.While image-based methods effectively capture appearance-related defects, these methods are limited in detecting structural and geometric anomalies, motivating multimodal approaches incorporating 3D information. However, existing methods mainly rely on local patch-level representations, which often lead to unstable reconstruction errors even in normal regions. To address this limitation, we propose GRC-Net, which integrates a global-attention MLP to enforce global representation consistency across patch embeddings with a stable reconstruction module to improve reconstruction stability. The proposed method captures holistic contextual information through a global token and suppresses reconstruction noise by minimizing discrepancies between original and predicted embeddings. Experiments on MVTec 3D-AD and Eyecandies demonstrate that GRC-Net consistently outperforms existing methods at both image and pixel levels. Qualitative results further demonstrate reduced reconstruction errors in normal regions and more distinct reconstruction differences between normal and anomalous regions.
Comments5 pages, 3 figures, Under Review
