SwinAD: Multi-stage feature reconstruction for unsupervised industrial anomaly detection

• URL: http://arxiv.org/abs/2607.14534v1
• Date: Thu, 16 Jul 2026 03:39:56 GMT
• Title: SwinAD: Multi-stage feature reconstruction for unsupervised industrial anomaly detection
• Abstract summary: SwinAD is a reconstruction-based framework for multi-class unsupervised anomaly detection.<n> hierarchical encoder provides semantically rich multi-scale features.<n> Feature diversity-preserving reconstruction framework maintains complementary reconstruction hypotheses.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Industrial anomaly detection aims to identify and localize defective regions without relying on exhaustive annotations of all possible defect types. Although recent unsupervised methods have achieved strong performance, most are primarily designed for single-class settings and often struggle in multi-class scenarios, where diverse normal patterns may lead to over-generalization and reduce the discriminative capability between normal and anomalous regions. In this paper, we propose SwinAD, a reconstruction-based framework for multi-class unsupervised anomaly detection that leverages a frozen pretrained Swin Transformer V2 encoder and a feature diversity-preserving reconstruction decoder. The hierarchical encoder provides semantically rich multi-scale features, while stage-wise bottleneck modules with dropout prevent trivial identity mapping and encourage robust reconstruction of normal patterns. To further improve localization, we introduce a feature diversity-preserving reconstruction framework that maintains complementary reconstruction hypotheses instead of relying on a single decoding branch. The discrepancies between encoder features and the two reconstructed features are then aggregated across multiple scales to produce the final anomaly map. Experiments conducted on three industrial anomaly detection benchmarks, including MVTec AD, VisA, and Real-IAD, demonstrate that SwinAD achieves competitive image-level performance and strong pixel-level localization accuracy, with particularly notable improvements in pixel-level AP and 1 on MVTec AD. These results indicate that combining hierarchical Swin features with diverse multi-scale reconstruction substantially improve pixel-level localization in multi-class unsupervised anomaly setting.

Related papers

• Point Cloud Diffusion with Global and Local Reconstruction for Instance-Level 3D Anomaly Detection [53.26344422459608]
  We propose textbfPCDiff, a point cloud diffusion framework for instance-level 3D anomaly generation and detection.<n>In the generation phase, an instance-level multi-modal attention is embedded into the generation framework, where anomalies are conditioned with texture gradient, image patch, text and mask.<n>In the detection phase, a joint local-global reconstruction algorithm is introduced to ensure local anomaly restoration and global geometric consistency.
  arXiv  Detail & Related papers  (2026-06-24T12:11:05Z)
• Text-Guided Multimodal Unified Industrial Anomaly Detection [71.95719669933312]
  We propose a unified multimodal industrial anomaly detection framework guided by text semantics.<n>The framework consists of two core modules: a Geometry-Aware Cross-Modal Mapper and an Object-Conditioned Textual Feature Adaptor.<n>Our method achieves state-of-the-art performance in classification and localization under unsupervised settings.
  arXiv  Detail & Related papers  (2026-04-24T13:21:22Z)
• Large-Scale Universal Defect Generation: Foundation Models and Datasets [51.150071370083026]
  We introduce UDG, a large-scale dataset of 300K normal-abnormal-mask-caption quadruplets spanning diverse domains.<n>We present UniDG, a universal defect generation foundation model that supports both reference-based defect generation and text instruction-based defect editing without per-category fine-tuning.<n>Experiments on MVTec-AD and VisA show that UniDG outperforms prior few-shot anomaly generation and image insertion/editing baselines in synthesis quality and downstream single- and multi-class anomaly detection/localization.
  arXiv  Detail & Related papers  (2026-04-10T03:21:17Z)
• Collaborative Reconstruction and Repair for Multi-class Industrial Anomaly Detection [37.057760207060554]
  Collaborative Reconstruction and Repair (CRR) transforms the reconstruction to repairation.<n>We implement feature-level random masking to ensure that the representations from decoder contain sufficient local information.<n>We train a segmentation network supervised by synthetic anomaly masks, thereby enhancing localization performance.
  arXiv  Detail & Related papers  (2025-12-12T09:24:23Z)
• ShortcutBreaker: Low-Rank Noisy Bottleneck with Global Perturbation Attention for Multi-Class Unsupervised Anomaly Detection [59.89803740308262]
  ShortcutBreaker is a novel unified feature-reconstruction framework for MUAD tasks.<n>It features two key innovations to address the issue of shortcuts.<n>The proposed method achieves a remarkable image-level AUROC of 99.8%, 98.9%, 90.6%, and 87.8% on four datasets.
  arXiv  Detail & Related papers  (2025-10-21T06:51:30Z)
• Generate Aligned Anomaly: Region-Guided Few-Shot Anomaly Image-Mask Pair Synthesis for Industrial Inspection [53.137651284042434]
  Anomaly inspection plays a vital role in industrial manufacturing, but the scarcity of anomaly samples limits the effectiveness of existing methods.<n>We propose Generate grained Anomaly (GAA), a region-guided, few-shot anomaly image-mask pair generation framework.<n>GAA generates realistic, diverse, and semantically aligned anomalies using only a small number of samples.
  arXiv  Detail & Related papers  (2025-07-13T12:56:59Z)
• Memory-Augmented Dual-Decoder Networks for Multi-Class Unsupervised Anomaly Detection [21.341378704908333]
  We propose a Memory-augmented Dual-Decoder Networks (MDD-Net)<n>This network includes two critical components: a Dual-Decoder Reverse Distillation Network (DRD-Net) and a Class-aware Memory Module (CMM)<n>By exploiting the discrepancy between features produced by two decoders, our approach refines anomaly scores beyond the conventional encoder-decoder comparison paradigm.
  arXiv  Detail & Related papers  (2025-04-21T06:29:04Z)
• Feature Purified Transformer With Cross-level Feature Guiding Decoder For Multi-class OOD and Anomaly Deteciton [22.59400036717643]
  We introduce the FUTUREG framework, which incorporates two innovative modules: the Feature Purification Module (FPM) and the CFG Decoder. The FPM constrains the normality boundary within the latent space to effectively filter out anomalous features. The CFG Decoder uses layer-wise encoder representations to guide the reconstruction of filtered features, preserving fine-grained details.
  arXiv  Detail & Related papers  (2024-04-30T16:45:51Z)
• Multi-feature Reconstruction Network using Crossed-mask Restoration for Unsupervised Industrial Anomaly Detection [4.742650815342744]
  Unsupervised anomaly detection is of great significance for quality inspection in industrial manufacturing. We propose a multi-feature reconstruction network, MFRNet, using crossed-mask restoration in this paper. Our method is highly competitive with or significantly outperforms other state-of-the-arts on four public available datasets and one self-made dataset.
  arXiv  Detail & Related papers  (2024-04-20T05:13:56Z)
• DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection [55.48770333927732]
  We propose a Difusion-based Anomaly Detection (DiAD) framework for multi-class anomaly detection. It consists of a pixel-space autoencoder, a latent-space Semantic-Guided (SG) network with a connection to the stable diffusion's denoising network, and a feature-space pre-trained feature extractor. Experiments on MVTec-AD and VisA datasets demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2023-12-11T18:38:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.