arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.16865cs.CV
收藏

配准是用于3D异常检测的强大旋转不变性学习器

Registration is a Powerful Rotation-Invariance Learner for 3D Anomaly Detection

South China University of Technology(华南理工大学);Guangdong University of Petrochemical Technology(广东石油化工学院);Singapore Management University(新加坡国立大学)

Yuyang Yu, Zhengwei Chen, Xuemiao Xu, Lei Zhang, Haoxin Yang, Yongwei Nie, Shengfeng He

AI总结:

针对点云3D异常检测中旋转不变性与局部判别不足的问题,提出融合点云配准与记忆库异常检测目标的特征提取框架,并在Anomaly-ShapeNet和Real3D-AD上验证其性能提升。

摘要:

点云数据中的3D异常检测对于工业质量控制至关重要,旨在以高可靠性识别结构缺陷。然而,当前基于记忆库的方法往往受到特征变换不一致和判别能力有限的困扰,尤其是在捕捉局部几何细节和实现旋转不变性方面。当配准失败时,这些局限性会变得更加明显,从而导致不可靠的检测结果。我们认为,点云配准不仅在对齐几何结构方面发挥重要作用,而且在引导特征提取朝旋转不变且具有局部判别性的表示发展方面也至关重要。为此,我们提出了一种配准诱导的旋转不变特征提取框架,该框架整合了点云配准和基于记忆的异常检测目标。我们的关键洞见是,这两项任务都依赖于对局部几何结构建模,并利用跨样本的特征相似性。通过将特征提取嵌入配准学习过程,我们的框架联合优化对齐和表示学习。这种整合使网络能够获得既对旋转具有鲁棒性、又对异常检测高度有效的特征。在Anomaly-ShapeNet和Real3D-AD数据集上进行的大量实验表明,我们的方法在有效性和泛化能力方面始终优于现有方法。

英文摘要:

3D anomaly detection in point-cloud data is critical for industrial quality control, aiming to identify structural defects with high reliability. However, current memory bank-based methods often suffer from inconsistent feature transformations and limited discriminative capacity, particularly in capturing local geometric details and achieving rotation invariance. These limitations become more pronounced when registration fails, leading to unreliable detection results. We argue that point-cloud registration plays an essential role not only in aligning geometric structures but also in guiding feature extraction toward rotation-invariant and locally discriminative representations. To this end, we propose a registration-induced, rotation-invariant feature extraction framework that integrates the objectives of point-cloud registration and memory-based anomaly detection. Our key insight is that both tasks rely on modeling local geometric structures and leveraging feature similarity across samples. By embedding feature extraction into the registration learning process, our framework jointly optimizes alignment and representation learning. This integration enables the network to acquire features that are both robust to rotations and highly effective for anomaly detection. Extensive experiments on the Anomaly-ShapeNet and Real3D-AD datasets demonstrate that our method consistently outperforms existing approaches in effectiveness and generalizability.

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。

↑