NexViTAD: Few-shot Unsupervised Cross-Domain Defect Detection via Vision Foundation Models and Multi-Task Learning

TL;DR

NexViTAD introduces a novel few-shot unsupervised cross-domain defect detection framework leveraging vision foundation models and multi-task learning to effectively address domain-shift challenges in industrial anomaly detection.

Contribution

The paper proposes a hierarchical adapter, shared subspace projection, and multi-task decoder architecture, advancing cross-domain defect detection with state-of-the-art performance.

Findings

Achieved 97.5% AUC on MVTec AD dataset.

Outperformed recent models in cross-domain defect detection.

Enhanced generalization through multi-task learning and feature fusion.

Abstract

This paper presents a novel few-shot cross-domain anomaly detection framework, Nexus Vision Transformer for Anomaly Detection (NexViTAD), based on vision foundation models, which effectively addresses domain-shift challenges in industrial anomaly detection through innovative shared subspace projection mechanisms and multi-task learning (MTL) module. The main innovations include: (1) a hierarchical adapter module that adaptively fuses complementary features from Hiera and DINO-v2 pre-trained models, constructing more robust feature representations; (2) a shared subspace projection strategy that enables effective cross-domain knowledge transfer through bottleneck dimension constraints and skip connection mechanisms; (3) a MTL Decoder architecture supports simultaneous processing of multiple source domains, significantly enhancing model generalization capabilities; (4) an anomaly score inference method based on Sinkhorn-K-means clustering, combined with Gaussian filtering and adaptive threshold processing for precise pixel level. Valuated on the MVTec AD dataset, NexViTAD delivers state-of-the-art performance with an AUC of 97.5%, AP of 70.4%, and PRO of 95.2% in the target domains, surpassing other recent models, marking a transformative advance in cross-domain defect detection.