mlsanityData to 14 Oct 2026

Reliability-Aware Prototype Calibration for Frozen Pose-Flow Video Anomaly Detection

arXiv 2606.20312 ↗First posted 18 Jun 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Pose-flow video anomaly detectors are attractive for one-class surveillance because they provide likelihood-based rankings for tracked skeleton windows. However, a single likelihood score may hide multimodal normal behavior and be sensitive to pose-observation noise. We study a frozen-detector setting in which the pose-flow backbone, cached skeleton tracks, and evaluation pipeline are fixed. Reliability-Aware Prototype Calibration (RPC) is a post-hoc score calibration method for this setting. It adds a standardized nearest-prototype deviation in the frozen latent space to the standardized flow score, and uses keypoint confidence only to gate this added geometric evidence. Thus, RPC preserves the original density signal while correcting the ranking with empirical normal-mode structure under pose reliability. Across two frozen pose-flow backbones and four datasets, RPC improves frame-level AUROC in all eight backbone-dataset pairs, with gains ranging from 0.34 to 4.49 percentage points and averaging 2.03 points. Ablation and reliability analyses show that prototype deviation is the main corrective signal, while reliability gating is most useful when pose observations are less trustworthy. These results suggest that lightweight post-hoc calibration can strengthen cached pose-flow systems when retraining or reproducing the full pose pipeline is impractical.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 27 of its 40 references that are in the map. First posted on arXiv 18 Jun 2026.

Builds on

27 of 40 references.

References · 27

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2021Generalized Out-of-Distribution Detection: A Surveycited by 802
  3. 2018Real-world Anomaly Detection in Surveillance Videoscited by 577
  4. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  5. 2017Future Frame Prediction for Anomaly Detection -- A New Baselinecited by 391
  6. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 221
  7. 2020Whole-Body Human Pose Estimation in the Wildcited by 178
  8. 2017PoseTrack: A Benchmark for Human Pose Estimation and Trackingcited by 173
  9. 2020Anomaly Detection in Video via Self-Supervised and Multi-Task Learningcited by 148
  10. 2019Graph Embedded Pose Clustering for Anomaly Detectioncited by 95
  11. 2021UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detectioncited by 93
  12. 2022Normalizing Flows for Human Pose Anomaly Detectioncited by 60
15 more
  1. 2023Generalized Video Anomaly Event Detection: Systematic Taxonomy and Comparison of Deep Modelscited by 59
  2. 2024Follow the Rules: Reasoning for Video Anomaly Detection with Large Language Modelscited by 57
  3. 2023Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectorscited by 37
  4. 2024Networking Systems for Video Anomaly Detection: A Tutorial and Surveycited by 21
  5. 2022Visual Anomaly Detection Via Partition Memory Bank Module and Error Estimationcited by 20
  6. 2024An Exploratory Study on Human-Centric Video Anomaly Detection through Variational Autoencoders and Trajectory Predictioncited by 19
  7. 2023Holistic Representation Learning for Multitask Trajectory Anomaly Detectioncited by 15
  8. 2025AnyAnomaly: Zero-Shot Customizable Video Anomaly Detection with LVLMcited by 15
  9. 2023Contracting Skeletal Kinematics for Human-Related Video Anomaly Detectioncited by 14
  10. 2024Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Modelcited by 14
  11. 2023Model Selection of Anomaly Detectors in the Absence of Labeled Validation Datacited by 8
  12. 2024VideoPatchCore: An Effective Method to Memorize Normality for Video Anomaly Detectioncited by 8
  13. 2024DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detectioncited by 7
  14. 2025Sequential keypoint density estimator: an overlooked baseline of skeleton-based video anomaly detectioncited by 7
  15. 2025Action Hints: Semantic Typicality and Context Uniqueness for Generalizable Skeleton-based Video Anomaly Detectioncited by 1

Built on by

Not cited yet.