PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization

By Thomas Defard, Aleksandr Setkov, Angélique Loesch, Romaric Audigier. ICPR Workshops, 2020

1,422 citations recorded by the source corpus.

We present a new framework for Patch Distribution Modeling, PaDiM, to concurrently detect and localize anomalies in images in a one-class learning setting. PaDiM makes use of a pretrained convolutional neural network (CNN) for patch embedding, and of multivariate Gaussian distributions to get a probabilistic representation of the normal class. It also exploits correlations between the different semantic levels of CNN to better localize anomalies. PaDiM outperforms current state-of-the-art approaches for both anomaly detection and localization on the MVTec AD and STC datasets. To match real-world visual industrial inspection, we extend the evaluation protocol to assess performance of anomaly localization algorithms on non-aligned dataset. The state-of-the-art performance and low complexity of PaDiM make it a good candidate for many industrial applications.

Subfields: Artificial Intelligence, Civil and Structural Engineering, Industrial and Manufacturing Engineering

View Paper: https://www.semanticscholar.org/paper/2e8d62277e40d465343e8dfb32ecc246f320540e

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

In Civil and Structural Engineering

In Industrial and Manufacturing Engineering

Citations in the corpus

WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation (2023, methodology): “Due to the scarcity of anomalies, the major focus has been on oneclass methods with many normal images [7, 8, 20, 48, 50, 51].”
DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detection (2022, methodology): “Most of the works downsampled the ground truth to 256 × 256 for faster computation, but some performed an extra 224 × 224 center crop [7, 8, 25].”
SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentation (2022, methodology): “For both setups, existing state-of-the-art methods for surface anomaly detection commonly leverage supervised representations pre-trained on ImageNet [16], either as feature extractors [14,33] or as initialization for fine-tuning on the target dataset [26,44].”
Towards Total Recall in Industrial Anomaly Detection (2021, methodology): “6% , and stride s = 3 an AUROC of 96 .”
Anomaly Detection via Reverse Distillation from One-Class Embedding (2022, methodology): “Recent pre-trained model based approaches achieve promising performance by extracting features from anomaly-free samples as a measurement [7, 8].”
DRÆM – A discriminatively trained reconstruction embedding for surface anomaly detection (2021, methodology): “Alternatively [20, 11] propose surface anomaly detection as identifying significant deviations from a Gaussian fitted to”
CFA: Coupled-Hypersphere-Based Feature Adaptation for Target-Oriented Anomaly Localization (2022, methodology): “Recently, [4, 5, 16] designed memory banks using pre-trained CNNs with large datasets such as ImageNet [6] and achieved state-of-the-art (SOTA) performance.”
AnomalyGPT: Detecting Industrial Anomalies using Large Vision-Language Models (2023, methodology): “Additionally, we compare our model with PaDiM [6] and JNLD [34] using the same unified setting.”
CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows (2021, methodology): “PaDiM [8] estimates train-time statistics i.”
SimpleNet: A Simple Network for Image Anomaly Detection and Localization (2023, methodology): “The technique to detect the various anomalies of interest has a broad set of applications in industrial inspection [3, 6].”
Deep Industrial Image Anomaly Detection: A Survey (2023, neutral): “PatchCore [69], SPADE [64], PaDim [74] conduct the ablation study on 16 normal training samples.”
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows (2021, methodology): “Most of them (Cohen and Hoshen 2020; Defard et al. 2020; Roth et al. 2021) use ResNet (He et al.”
A Unified Model for Multi-class Anomaly Detection (2022, background): “US [3] PSVDD [15] PaDiM [5] CutPaste [10] FCDD [11] MKD [12] DRAEM [16] Ours FLOPs(G) 60.”
Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection (2021, methodology): “Anomaly detection methods can be classified into: dictionary learning methods [7–9, 14, 40, 55], change detection frameworks [11, 26, 38, 48], probability-based methods [1, 2, 16, 23, 29, 44, 45, 56, 61, 74], distance-based models [6, 10, 25, 27, 50, 51,53,57,58,63,65,68,70], and reconstruction-base”
Student-Teacher Feature Pyramid Matching for Anomaly Detection (2021, methodology): “We then consider the task of pixel-level anomaly detection and compare our method with the counterparts including Patch-SVDD [43], PaMiD [13], etc .”
EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies (2023, methodology): “[16] Thomas Defard, Aleksandr Setkov, Angelique Loesch, and Romaric Audigier.”
Surface Defect Detection Methods for Industrial Products: A Review (2021, methodology): “Typical algorithms mainly include SPADE [54], PaDIM [55], PatchCore [56], etc.”