mlsanityData to 14 Oct 2026

Normalizing Flows for Human Pose Anomaly Detection

arXiv 2211.10946 ↗First posted 20 Nov 2022cs.CV cs.AI eess.IVcited by 60

Not on the map. Why?

Abstract (arXiv)

Video anomaly detection is an ill-posed problem because it relies on many parameters such as appearance, pose, camera angle, background, and more. We distill the problem to anomaly detection of human pose, thus decreasing the risk of nuisance parameters such as appearance affecting the result. Focusing on pose alone also has the side benefit of reducing bias against distinct minority groups. Our model works directly on human pose graph sequences and is exceptionally lightweight (~1K parameters), capable of running on any machine able to run the pose estimation with negligible additional resources. We leverage the highly compact pose representation in a normalizing flows framework, which we extend to tackle the unique characteristics of spatio-temporal pose data and show its advantages in this use case. The algorithm is quite general and can handle training data of only normal examples as well as a supervised setting that consists of labeled normal and abnormal examples. We report state-of-the-art results on two anomaly detection benchmarks - the unsupervised ShanghaiTech dataset and the recent supervised UBnormal dataset.

Key facts. Cited by 60 papers in the map (data through 14 Oct 2026). Builds on 39 of its 49 references that are in the map. First posted on arXiv 20 Nov 2022.

Builds on

39 of 49 references.

References · 39

  1. 2016Density estimation using Real NVPcited by 2,042
  2. 2021Is Space-Time Attention All You Need for Video Understanding?cited by 1,620
  3. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,545
  4. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,209
  5. 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,136
  6. 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,071
  7. 2014NICE: Non-linear Independent Components Estimationcited by 1,058
  8. 2018Real-world Anomaly Detection in Surveillance Videoscited by 577
  9. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 564
  10. 2019Using Self-Supervised Learning Can Improve Model Robustness and Uncertaintycited by 514
  11. 2017Future Frame Prediction for Anomaly Detection -- A New Baselinecited by 391
  12. 2020Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognitioncited by 343
27 more
  1. 2020Learning Memory-guided Normality for Anomaly Detectioncited by 278
  2. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 221
  3. 2018CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmarkcited by 214
  4. 2020Anomaly Detection in Video via Self-Supervised and Multi-Task Learningcited by 148
  5. 2021A Hybrid Video Anomaly Detection Framework via Memory-Augmented Flow Reconstruction and Flow-Guided Frame Predictioncited by 129
  6. 2021Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detectioncited by 119
  7. 2019Anomaly Detection in Video Sequence with Appearance-Motion Correspondencecited by 118
  8. 2020Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biasescited by 107
  9. 2019Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videoscited by 104
  10. 2019Graph Normalizing Flowscited by 104
  11. 2020Deep Nearest Neighbor Anomaly Detectioncited by 104
  12. 2022Generative Cooperative Learning for Unsupervised Video Anomaly Detectioncited by 102
  13. 2019Graph Embedded Pose Clustering for Anomaly Detectioncited by 95
  14. 2021UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detectioncited by 93
  15. 2019Street Scene: A new dataset and evaluation protocol for video anomaly detectioncited by 90
  16. 2020Saving Face: Investigating the Ethical Concerns of Facial Recognition Auditingcited by 89
  17. 2020A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Videocited by 86
  18. 2018Pose Flow: Efficient Online Pose Trackingcited by 80
  19. 2019Semi-Supervised Learning with Normalizing Flowscited by 76
  20. 2022Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzlescited by 62
  21. 2019Multi-timescale Trajectory Prediction for Abnormal Human Activity Detectioncited by 59
  22. 2018Racial categories in machine learningcited by 48
  23. 2022SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detectioncited by 45
  24. 2019Deep Learning for Face Recognition: Pride or Prejudiced?cited by 25
  25. 2020Unsupervised Video Anomaly Detection via Normalizing Flows with Implicit Latent Featurescited by 23
  26. 2021Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?cited by 16
  27. 2021Graph-based Normalizing Flow for Human Motion Generation and Reconstructioncited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 6; 2024: 18; 2025: 18; 2026: 18.

60 papers.

Sort:

By area:

  1. 2024Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Modelscited by 57
  2. 2024Advancing Video Anomaly Detection: A Concise Review and a New Datasetcited by 49
  3. 2023Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectorscited by 37
  4. 2024NRDF: Neural Riemannian Distance Fields for Learning Articulated Pose Priorscited by 27
  5. 2024Deep Learning for Video Anomaly Detection: A Reviewcited by 27
  6. 2023BatchNorm-based Weakly Supervised Video Anomaly Detectioncited by 22
  7. 2024MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detectioncited by 22
  8. 2023Understanding the Challenges and Opportunities of Pose-based Anomaly Detectioncited by 15
  9. 2025Anomize: Better Open Vocabulary Video Anomaly Detectioncited by 13
  10. 2024Privacy-Preserving Video Anomaly Detection: A Surveycited by 11
  11. 2024Beyond Walking: A Large-Scale Image-Text Benchmark for Text-based Person Anomaly Searchcited by 11
  12. 2024VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detectioncited by 8
38 more
  1. 2026No Need For Real Anomaly: MLLM Empowered Zero-Shot Video Anomaly Detectioncited by 8
  2. 2026RC-NF: Robot-Conditioned Normalizing Flow for Real-Time Anomaly Detection in Robotic Manipulationcited by 8
  3. 2024Evaluating the Effectiveness of Video Anomaly Detection in the Wild: Online Learning and Inference for Real-world Deploymentcited by 7
  4. 2024DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detectioncited by 7
  5. 2025Sequential keypoint density estimator: an overlooked baseline of skeleton-based video anomaly detectioncited by 7
  6. 2024Towards Adaptive Human-centric Video Anomaly Detection: A Comprehensive Framework and A New Benchmarkcited by 6
  7. 2024Human-Centric Video Anomaly Detection Through Spatio-Temporal Pose Tokenization and Transformercited by 6
  8. 2024Vision-Language Models Assisted Unsupervised Video Anomaly Detectioncited by 6
  9. 2024Frequency-Guided Diffusion Model with Perturbation Training for Skeleton-Based Video Anomaly Detectioncited by 6
  10. 2025Shopformer: Transformer-Based Framework for Detecting Shoplifting via Human Posecited by 6
  11. 2024Interleaving One-Class and Weakly-Supervised Models with Adaptive Thresholding for Unsupervised Video Anomaly Detectioncited by 5
  12. 2025Exploring Pose-Based Anomaly Detection for Retail Security: A Real-World Shoplifting Dataset and Benchmarkcited by 5
  13. 2025MoFM: A Large-Scale Human Motion Foundation Modelcited by 5
  14. 2025A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understandingcited by 5
  15. 2025Deep Incomplete Multi-view Clustering with Distribution Dual-Consistency Recovery Guidancecited by 4
  16. 2025Autoregressive Denoising Score Matching is a Good Video Anomaly Detectorcited by 4
  17. 2025The Evolution of Video Anomaly Detection: A Unified Framework from DNN to MLLMcited by 4
  18. 2025A Survey on Video Anomaly Detection via Deep Learning: Human, Vehicle, and Environmentcited by 4
  19. 2024IPAD: Industrial Process Anomaly Detection Datasetcited by 3
  20. 2024Unveiling Context-Related Anomalies: Knowledge Graph Empowered Decoupling of Scene and Action for Human-Related Video Anomaly Detectioncited by 3
  21. 2025Detecting Contextual Anomalies by Discovering Consistent Spatial Regionscited by 3
  22. 2025ALFred: An Active Learning Framework for Real-world Semi-supervised Anomaly Detection with Adaptive Thresholdscited by 3
  23. 2026From Offline to Periodic Adaptation for Pose-Based Shoplifting Detection in Real-world Retail Securitycited by 3
  24. 2026Bounding-Box Trajectories Matter for Video Anomaly Detectioncited by 3
  25. 2024Frequency-Guided Multi-Level Human Action Anomaly Detection with Normalizing Flowscited by 2
  26. 2026Are Multimodal LLMs Ready for Surveillance? A Reality Check on Zero-Shot Anomaly Detection in the Wildcited by 2
  27. 2023Divide and Conquer in Video Anomaly Detection: A Comprehensive Review and New Approachcited by 1
  28. 2023Human Kinematics-inspired Skeleton-based Video Anomaly Detectioncited by 1
  29. 2025Efficient Vision Language Model Fine-tuning for Text-based Person Anomaly Searchcited by 1
  30. 2025Few-shot Human Action Anomaly Detection via a Unified Contrastive Learning Frameworkcited by 1
  31. 2025Action Hints: Semantic Typicality and Context Uniqueness for Generalizable Skeleton-based Video Anomaly Detectioncited by 1
  32. 2026VADMamba++: Efficient Video Anomaly Detection via Hybrid Modeling in Grayscale Spacecited by 1
  33. 2026From Frames to Events: Rethinking Evaluation in Human-Centric Video Anomaly Detectioncited by 1
  34. 2023A brief introduction to a framework named Multilevel Guidance-Exploration Networkcited by 0
  35. 2025Video Anomaly Detection with Contours -- A Studycited by 0
  36. 2025Action Dubber: Timing Audible Actions via Inflectional Flowcited by 0
  37. 20253D Human-Human Interaction Anomaly Detectioncited by 0
  38. 2026BUSSARD: Normalizing Flows for Bijective Universal Scene-Specific Anomalous Relationship Detectioncited by 0

50 most cited of 60 All 60 (JSON)