Unsupervised Multi-View Visual Anomaly Detection via Progressive Homography-Guided Alignment

TL;DR

This paper introduces ViewSense-AD, a framework for unsupervised multi-view visual anomaly detection that aligns features across views using homography, improving robustness to viewpoint changes and achieving state-of-the-art results.

Contribution

It proposes a novel multi-view alignment module with homography, integrated into a diffusion model, to learn viewpoint-invariant representations for anomaly detection.

Findings

Outperforms existing methods on RealIAD and MANTA datasets

Achieves state-of-the-art accuracy in pixel, view, and sample-level detection

Demonstrates robustness to large viewpoint shifts and complex textures

Abstract

Unsupervised visual anomaly detection from multi-view images presents a significant challenge: distinguishing genuine defects from benign appearance variations caused by viewpoint changes. Existing methods, often designed for single-view inputs, treat multiple views as a disconnected set of images, leading to inconsistent feature representations and a high false-positive rate. To address this, we introduce ViewSense-AD (VSAD), a novel framework that learns viewpoint-invariant representations by explicitly modeling geometric consistency across views. At its core is our Multi-View Alignment Module (MVAM), which leverages homography to project and align corresponding feature regions between neighboring views. We integrate MVAM into a View-Align Latent Diffusion Model (VALDM), enabling progressive and multi-stage alignment during the denoising process. This allows the model to build a coherent and holistic understanding of the object's surface from coarse to fine scales. Furthermore, a lightweight Fusion Refiner Module (FRM) enhances the global consistency of the aligned features, suppressing noise and improving discriminative power. Anomaly detection is performed by comparing multi-level features from the diffusion model against a learned memory bank of normal prototypes. Extensive experiments on the challenging RealIAD and MANTA datasets demonstrate that VSAD sets a new state-of-the-art, significantly outperforming existing methods in pixel, view, and sample-level visual anomaly proving its robustness to large viewpoint shifts and complex textures.