mlsanityData to 7 Oct 2026

From Offline to Periodic Adaptation for Pose-Based Shoplifting Detection in Real-world Retail Security

arXiv 2603.04723 ↗First posted 5 Mar 2026cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

Shoplifting is a growing operational and economic challenge for retailers, with incidents rising and losses increasing despite extensive video surveillance. Continuous human monitoring is infeasible, motivating automated, privacy-preserving, and resource-aware detection solutions. In this paper, we cast shoplifting detection as a pose-based, unsupervised video anomaly detection problem and introduce a periodic adaptation framework designed for on-site Internet of Things (IoT) deployment. Our approach enables edge devices in smart retail environments to adapt from streaming, unlabeled data, supporting scalable and low-latency anomaly detection across distributed camera networks. To support reproducibility, we introduce RetailS, a new large-scale real-world shoplifting dataset collected from a retail store under multi-day, multi-camera conditions, capturing unbiased shoplifting behavior in realistic IoT settings. For deployable operation, thresholds are selected using both F1 and H_PRS scores, the harmonic mean of precision, recall, and specificity, during data filtering and training. In periodic adaptation experiments, our framework consistently outperformed offline baselines on AUC-ROC and AUC-PR in 91.6% of evaluations, with each training update completing in under 30 minutes on edge-grade hardware, demonstrating the feasibility and reliability of our solution for IoT-enabled smart retail deployment.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 39 of its 65 references that are in the map. First posted on arXiv 5 Mar 2026.

Builds on

39 of 65 references.

References · 39

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  3. 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 812
  4. 2018Real-world Anomaly Detection in Surveillance Videoscited by 570
  5. 2017Future Frame Prediction for Anomaly Detection -- A New Baselinecited by 391
  6. 2021Revisiting Skeleton-based Action Recognitioncited by 256
  7. 2020Anomaly Detection in Video via Self-Supervised and Multi-Task Learningcited by 147
  8. 2019Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videoscited by 103
  9. 2022Generative Cooperative Learning for Unsupervised Video Anomaly Detectioncited by 102
  10. 2019Graph Embedded Pose Clustering for Anomaly Detectioncited by 95
  11. 2020A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Videocited by 85
  12. 2020Rethinking the Heatmap Regression for Bottom-up Human Pose Estimationcited by 69
27 more
  1. 2022Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzlescited by 62
  2. 2019Multi-timescale Trajectory Prediction for Abnormal Human Activity Detectioncited by 59
  3. 2022Normalizing Flows for Human Pose Anomaly Detectioncited by 59
  4. 2024Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Modelscited by 55
  5. 2019Deep Contextual Attention for Human-Object Interaction Detectioncited by 54
  6. 2020Continual Learning for Anomaly Detection in Surveillance Videoscited by 49
  7. 2022SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detectioncited by 45
  8. 2021A Hierarchical Spatio-Temporal Graph Convolutional Neural Network for Anomaly Detection in Videoscited by 28
  9. 2020Online Anomaly Detection in Surveillance Videos with Asymptotic Bounds on False Alarm Ratecited by 22
  10. 2021Regularity Learning via Explicit Distribution Modeling for Skeletal Video Anomaly Detectioncited by 22
  11. 2024An Exploratory Study on Human-Centric Video Anomaly Detection through Variational Autoencoders and Trajectory Predictioncited by 19
  12. 2022CHAD: Charlotte Anomaly Datasetcited by 18
  13. 2023Ancilia: Scalable Intelligent Video Surveillance for the Artificial Intelligence of Thingscited by 17
  14. 2023Understanding the Challenges and Opportunities of Pose-based Anomaly Detectioncited by 14
  15. 2020Anomaly Detection with Domain Adaptationcited by 13
  16. 2024Context-aware Video Anomaly Detection in Long-Term Datasetscited by 11
  17. 2024Evaluating the Effectiveness of Video Anomaly Detection in the Wild: Online Learning and Inference for Real-world Deploymentcited by 7
  18. 2022Multi-Task Learning based Video Anomaly Detection with Attentioncited by 6
  19. 2024Dynamic Distinction Learning: Adaptive Pseudo Anomalies for Video Anomaly Detectioncited by 6
  20. 2024Quantifying Influencer Impact on Affective Polarizationcited by 6
  21. 2024Human-Centric Video Anomaly Detection Through Spatio-Temporal Pose Tokenization and Transformercited by 6
  22. 2025Shopformer: Transformer-Based Framework for Detecting Shoplifting via Human Posecited by 6
  23. 2023Multi-level Memory-augmented Appearance-Motion Correspondence Framework for Video Anomaly Detectioncited by 5
  24. 2023From Lab to Field: Real-World Evaluation of an AI-Driven Smart Video Solution to Enhance Community Safetycited by 5
  25. 2025Exploring Pose-Based Anomaly Detection for Retail Security: A Real-World Shoplifting Dataset and Benchmarkcited by 5
  26. 2025Detecting Zero-Day Web Attacks with an Ensemble of LSTM, GRU, and Stacked Autoencoderscited by 4
  27. 2025Adversarially-Refined VQ-GAN with Dense Motion Tokenization for Spatio-Temporal Heatmapscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inferencecited by 3
  2. 2026From Frames to Events: Rethinking Evaluation in Human-Centric Video Anomaly Detectioncited by 1
  3. 2026VQ-VAD: Vector-quantized Motion Representation Learning for Human-centric Video Anomaly Detectioncited by 0