Unsupervised Multi-View Visual Anomaly Detection via Progressive Homography-Guided Alignment
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (4) · 3065-3073
DOI: 10.1609/aaai.v40i4.37299 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Xintao Chen und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).
Abstract (Originalsprache)
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 detection, proving its robustness to large viewpoint shifts and complex textures.
Zitieren
APA
Chen, X., Xu, X., Zheng, B., Liu, Y., & Wu, Y. (2026). Unsupervised Multi-View Visual Anomaly Detection via Progressive Homography-Guided Alignment. Proceedings of the AAAI Conference on Artificial Intelligence, 40(4), 3065-3073. https://doi.org/10.1609/aaai.v40i4.37299
Vancouver
Chen X, Xu X, Zheng B, Liu Y, Wu Y. Unsupervised Multi-View Visual Anomaly Detection via Progressive Homography-Guided Alignment. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(4):3065-3073. doi:10.1609/aaai.v40i4.37299
BibTeX
@article{Chen2026,
author = {Chen, Xintao and Xu, Xiaohao and Zheng, Bozhong and Liu, Yun and Wu, Yingna},
title = {Unsupervised Multi-View Visual Anomaly Detection via Progressive Homography-Guided Alignment},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {4},
pages = {3065--3073},
doi = {10.1609/aaai.v40i4.37299}
}
EUVIMED