Memoryless Multimodal Anomaly Detection via Student-Teacher Network and Signed Distance Learning

TL;DR

This paper introduces MDSS, a memoryless multimodal anomaly detection method using a student-teacher network and signed distance learning, achieving stable and superior detection performance without extra memory overhead.

Contribution

The paper proposes a novel memoryless approach for multimodal anomaly detection combining a student-teacher network and signed distance learning, reducing memory costs and improving stability.

Findings

MDSS outperforms baseline methods in anomaly detection accuracy.

MDSS is more stable than memory bank-based methods.

The approach effectively integrates RGB and 3D point cloud data.

Abstract

Unsupervised anomaly detection is a challenging computer vision task, in which 2D-based anomaly detection methods have been extensively studied. However, multimodal anomaly detection based on RGB images and 3D point clouds requires further investigation. The existing methods are mainly inspired by memory bank based methods commonly used in 2D-based anomaly detection, which may cost extra memory for storing mutimodal features. In present study, a novel memoryless method MDSS is proposed for multimodal anomaly detection, which employs a light-weighted student-teacher network and a signed distance function to learn from RGB images and 3D point clouds respectively, and complements the anomaly information from the two modalities. Specifically, a student-teacher network is trained with normal RGB images and masks generated from point clouds by a dynamic loss, and the anomaly score map could be obtained from the discrepancy between the output of student and teacher. Furthermore, the signed distance function learns from normal point clouds to predict the signed distances between points and surface, and the obtained signed distances are used to generate anomaly score map. Subsequently, the anomaly score maps are aligned to generate the final anomaly score map for detection. The experimental results indicate that MDSS is comparable but more stable than the SOTA memory bank based method Shape-guided, and furthermore performs better than other baseline methods.