MATCH: Flow Matching for Multi-View Anomaly Detection

2026-06-23

AI Summary

This work addresses the challenge of anomaly detection for complex industrial objects under multi-view imaging by introducing flow matching into this domain for the first time. Leveraging its ordinary differential equation (ODE) formulation, the method efficiently estimates data likelihood and constructs an anomaly scoring mechanism that avoids computationally expensive divergence terms. It enables flexible mapping of multi-scale features to a standard normal distribution and achieves feature alignment and transformation across multiple views. Evaluated on the Real-IAD and MANTA-Tiny datasets, the proposed approach attains state-of-the-art performance in both image-level and pixel-level anomaly detection and segmentation tasks, while supporting real-time inference on consumer-grade hardware.

Abstract

Detecting anomalies in industrial objects is an important topic for increasing production efficiency. More complex objects often require the analysis of several view points, which has led to the field of multi-view anomaly detection. We present MATCH, the first multi-view anomaly detection method based on Flow Matching (FM). With the ODE formulation of Flow Matching, we can estimate likelihoods and thereby derive an anomaly score to detect anomalies in multi-view image data at object, image, and pixel-level. The architectural flexibility of FM models allows us to efficiently transform features of different spatial sizes to the normal distribution. We evaluate thoroughly on the already established Real-IAD data set and are also the first to provide a comprehensive evaluation of popular anomaly detection methods for the MANTA-Tiny data set. MATCH achieves state-of-the-art performance in both anomaly detection and segmentation, all while running on consumer-level hardware. By omitting the costly divergence term needed for likelihood estimation, we ensure that MATCH is usable in real-time production scenarios. Lastly, several ablation studies are conducted to validate the methodological choices.

Problem

• multi-view anomaly detection
• industrial anomaly detection
• anomaly segmentation
• likelihood estimation
• real-time anomaly detection

Innovation

• Flow Matching
• Multi-View Anomaly Detection
• Likelihood Estimation
• Real-Time Inference
• Industrial Inspection