DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection

15 Mar 2023  ·  HUI ZHANG, Zheng Wang, Zuxuan Wu, Yu-Gang Jiang ·

Anomaly detection has garnered extensive applications in real industrial manufacturing due to its remarkable effectiveness and efficiency. However, previous generative-based models have been limited by suboptimal reconstruction quality, hampering their overall performance. A fundamental enhancement lies in our reformulation of the reconstruction process using a diffusion model into a noise-to-norm paradigm. Here, anomalous regions are perturbed with Gaussian noise and reconstructed as normal, overcoming the limitations of previous models by facilitating anomaly-free restoration. Additionally, we propose a rapid one-step denoising paradigm, significantly faster than the traditional iterative denoising in diffusion models. Furthermore, the introduction of the norm-guided paradigm elevates the accuracy and fidelity of reconstructions. The segmentation sub-network predicts pixel-level anomaly scores using the input image and its anomaly-free restoration. Comprehensive evaluations on four standard and challenging benchmarks reveal that DiffusionAD outperforms current state-of-the-art approaches, demonstrating the effectiveness and broad applicability of the proposed pipeline.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Anomaly Detection DAGM2007 DiffusionAD Detection AUROC 99.6 # 1
Anomaly Detection MVTec AD DiffusionAD Detection AUROC 99.7 # 5
Segmentation AUROC 98.7 # 13
Segmentation AUPRO 95.7 # 10
Segmentation AP 76.1 # 5
FPS 23.5 # 13
Anomaly Detection VisA DiffusionAD Detection AUROC 98.8 # 2
Segmentation AUPRO (until 30% FPR) 96.0 # 1
Segmentation AUROC 98.9 # 1
Segmentation AUPRO 96.0 # 1

Methods