VT-ADL: A Vision Transformer Network for Image Anomaly Detection and Localization

By Pankaj Kumar Mishra, Riccardo Verk, Daniele Fornasier, Claudio Piciarelli, Gian Luca Foresti. International Symposium on Industrial Electronics, 2021

523 citations recorded by the source corpus.

We present a transformer-based image anomaly detection and localization network. Our proposed model is a combination of a reconstruction-based approach and patch embedding. The use of transformer networks helps preserving the spatial information of the embedded patches, which is later processed by a Gaussian mixture density network to localize the anomalous areas. In addition, we also publish BTAD, a real-world industrial anomaly dataset. Our results are compared with other state-of-the-art algorithms using publicly available datasets like MNIST and MVTec.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/913d43b69fd4ddee2ff64d3a1e6402d3787e2b7e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Nguyen H. Tran
2. Xin Lu
3. Rui Zhao
4. Ye Zheng
5. Onkar Dabeer
6. Kilian Batzner
7. Xinyi Le
8. Anh Tuan Nguyen

In Computer Vision and Pattern Recognition

1. Xiaogang Wang
2. Yi Yang
3. Luc Van Gool
4. Yujun Shen
5. Ming-Hsuan Yang
6. Dacheng Tao
7. Trevor Darrell
8. Xin Lu

In Radiology, Nuclear Medicine and Imaging

1. Kilian Batzner
2. Lars Heckler
3. Avinash Ravichandran
4. Rebecca König

Citations in the corpus

• AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection (2023, unknown): “(Jezek et al., 2021), BTAD (Mishra et al., 2021), SDD (Tabernik et al., 2020), DAGM (Wieler & Hahn, 2007), and DTD-Synthetic (Aota et al., 2023).”
• FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows (2021, background): “BTAD BeanTech Anomaly Detection dataset (Mishra et al. 2021) has 3 categories industrial products with 2540 images.”
• Revisiting Reverse Distillation for Anomaly Detection (2023, methodology): “Anomaly detection and localization results in terms of AUROC at image-level/pixel-level/PRO on the BTAD dataset [22].”
• Deep Industrial Image Anomaly Detection: A Survey (2023, unknown): “(3) Transformer Model VT-ADL [107] L2, SSIM, Log-Likelihood The paper proposes a transformer-based framework to reconstruct images and detects anomalies.”
• A Unified Model for Multi-class Anomaly Detection (2022, methodology): “VT-ADL [28] and AnoVit [50] both apply transformer encoder to reconstruct images.”
• WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation (2023, background): “Additionally, Vision Transformer (ViT) have recently shown its potential in visual inspection [9, 25].”
• EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies (2023, methodology): “Recently, several new industrial anomaly detection datasets have been introduced [8, 11, 25, 27, 37, 69].”