Awesome

Awesome-XAD

Paper and dataset collection for the paper.

Explainable 2D Anomaly Detection Methods

[Image+Video]

- Attention-based

• [AD-FactorVAE] Towards visually explaining variational autoencoders. | CVPR 2020 | [pdf] [code]
• [CAVGA] Attention guided anomaly localization in images. | ECCV 2020 | [pdf]
• [SSPCAB] Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection. | CVPR 2022 | [pdf] [code]

- Generative-model-based

• [LSAND] Latent space autoregression for novelty detection. | CVPR 2019 | [pdf] [code]
• [CFLOW-AD] CFLOW-AD: Real-time unsupervised anomaly detection with localization via conditional normalizing flows. | WACV 2022 | [pdf] [code]

[Image]

- Attention-based

• [Gradcon] Backpropagated gradient representations for anomaly detection. | ECCV 2020 | [pdf] [code]
• [FCCD] Explainable deep one-class classification. | ICLR 2021 | [pdf] [code]

- Input-perturbation-based

• [ODIN] Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks. | ICLR 2018 | [pdf] [code]
• [Mahalanobis] A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks. | NIPS 2018 | [pdf] [code]
• [Generalized-ODIN] Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Data. | CVPR 2020 | [pdf] [code]
• [SLA2P] Self-supervision meets adversarial perturbation: A novel framework for anomaly detection. | CIKM 2022 | [pdf] [code]

- Generative-model-based

• [AnoGAN] Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery. | IPMI 2017 | [pdf] [code]
• [ALAD] Adversarially Learned Anomaly Detection. | ICDM 2018 | [pdf] [code]
• [f-AnoGAN] f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks. | Medical Image Analysis 2019 | [pdf] [code]
• [Genomics-OOD] Likelihood Ratios for Out-of-Distribution Detection. | NeurIPS 2019 | [pdf] [code]
• [Likelihood-Regret] Likelihood Regret: An Out-of-Distribution Detection Score For Variational Auto-encoder. | NeurIPS 2020 | [pdf] [code]
• [FastFlow] FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows. | arXiv 2021 | [pdf] [code]
• [AnoDDPM] AnoDDPM: Anomaly Detection with Denoising Diffusion Probabilistic Models using Simplex Noise. | CVPRW 2022 | [pdf] [code]
• [Diffusion-anomaly] Diffusion Models for Medical Anomaly Detection. | MICCAI 2022 | [pdf] [code]
• [DDPM] Fast unsupervised brain anomaly detection and segmentation with diffusion models. | MICCAI 2022 | [pdf]
• [DiAD] DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection. | AAAI 2024 | [pdf] [code]

- Foundation-model-based

• [WinCLIP] WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation. | CVPR 2023 | [pdf] [code]
• [CLIP-AD] CLIP-AD: A Language-Guided Staged Dual-Path Model for Zero-shot Anomaly Detection. | arXiv 2023 | [pdf] [code]
• [AnomalyCLIP] AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection. | ICLR 2024 | [pdf] [code]
• [SAA+] Segment Any Anomaly without Training via Hybrid Prompt Regularization. | arXiv 2023 | [pdf] [code]
• [AnomalyGPT] AnomalyGPT: Detecting Industrial Anomalies Using Large Vision-Language Models. | AAAI 2024 | [pdf] [code]
• [Myriad] Myriad: Large Multimodal Model by Applying Vision Experts for Industrial Anomaly Detection. | arXiv 2023 | [pdf] [code]
• [GPT-4V] Towards Generic Anomaly Detection and Understanding: Large-scale Visual-linguistic Model (GPT-4V) Takes the Lead. | arXiv 2023 | [pdf] [code]
• [GPT-4V-AD] Exploring Grounding Potential of VQA-oriented GPT-4V for Zero-shot Anomaly Detection. | arXiv 2023 | [pdf] [code]

[Video]

- Attention-based

• [Self-trained-DOR] Self-trained Deep Ordinal Regression for End-to-End Video Anomaly Detection. | CVPR 2020 | [pdf]
• [DSA] Dance with self-attention: A new look of conditional random fields on anomaly detection in videos. | ICCV 2021 | [pdf]

- Reasoning-based

• [Scene-Graph] Scene graphs for interpretable video anomaly classification. | NIPSW 2018 | [pdf]
• [CTR] Learning causal temporal relation and feature discrimination for anomaly detection. | TIP 2021 | [pdf]
• [Interpretable] Towards interpretable video anomaly detection. | WACV 2023 | [pdf]
• [VADor-w-LSTC] Video Anomaly Detection and Explanation via Large Language Models. | arXiv 2024 | [pdf]

- Intrinsic interpretable

• [JDR] Joint detection and recounting of abnormal events by learning deep generic knowledge. | ICCV 2017 | [pdf]
• [XMAN] X-MAN: Explaining multiple sources of anomalies in video. | CVPRW 2021 | [pdf]
• [VQU-Net] Discrete neural representations for explainable anomaly detection. | WACV 2022 | [pdf]
• [AI-VAD] Attribute-based Representations for Accurate and Interpretable Video Anomaly Detection. | arXiv 2022 | [pdf] [code]
• [EVAL] Eval: Explainable video anomaly localization. | arXiv 2022 | [pdf]

Datasets

[Image]

- Classification datasets for semantic anomaly detection and OOD detection

[2009] CIFAR-10, [2009] CIFAR-100, [2011] Texture, [2011] SVHN, [2012] MNIST, [2003] Caltech-101, [2015] LSUN, [2015] iSUN, [2015] CelebA, [2017] Fashion-MNIST, [2017] CURE-TSR

- Sensory anomaly detection datasets

[2007] DAGM, [2012] BRATS, [2015] UK-Biobank, [2017] MSLUB, [2019] WMH, [2019] Fishyscapes, [2019] CheXpert, [2019] LAG, [2019] MVTec AD, [2020] MTD, [2020] SDD, [2021] BTAD, [2021] MPDD, [2021] MedMNIST, [2021] KSSD2, [2021] RoadAnomaly21, [2022] VisA, [2022] MVTec LOCO AD

[Video]

[2008] Subway, [2009] UMN, [2010] UCSD-Ped, [2013] CUHK-Avenue, [2018] UCF-Crime, [2018] ShanghaiTech-Campus, [2019] Street-Scene, [2020] XD-Violence, [2021] X-MAN, [2021] TAD, [2022] UBnormal, [2023] NWPU-Campus