Action Hints: Semantic Typicality and Context Uniqueness for Generalizable Skeleton-based Video Anomaly Detection
Not on the map. Why?
Abstract (arXiv)
Zero-Shot Video Anomaly Detection (ZS-VAD) requires temporally localizing anomalies without target domain training data, which is a crucial task due to various practical concerns, e.g., data privacy or new surveillance deployments. Skeleton-based approach has inherent generalizable advantages in achieving ZS-VAD as it eliminates domain disparities both in background and human appearance. However, existing methods only learn low-level skeleton representation and rely on the domain-limited normality boundary, which cannot generalize well to new scenes with different normal and abnormal behavior patterns. In this paper, we propose a novel zero-shot video anomaly detection framework, unlocking the potential of skeleton data via action typicality and uniqueness learning. Firstly, we introduce a language-guided semantic typicality modeling module that projects skeleton snippets into action semantic space and distills LLM's knowledge of typical normal and abnormal behaviors during training. Secondly, we propose a test-time context uniqueness analysis module to finely analyze the spatio-temporal differences between skeleton snippets and then derive scene-adaptive boundaries. Without using any training samples from the target domain, our method achieves state-of-the-art results against skeleton-based methods on four large-scale VAD datasets: ShanghaiTech, UBnormal, NWPU, and UCF-Crime, featuring over 100 unseen surveillance scenes.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 43 of its 54 references that are in the map. First posted on arXiv 14 Sep 2025.
Builds on
43 of 54 references.
References · 43
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
- 2025Qwen2.5-VL Technical Reportcited by 5,587
- 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
- 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,270
- 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,545
- 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,209
- 2023ImageBind: One Embedding Space To Bind Them Allcited by 1,175
- 2018OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,162
- 2018Real-world Anomaly Detection in Surveillance Videoscited by 577
- 2017Future Frame Prediction for Anomaly Detection -- A New Baselinecited by 391
- 2021Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognitioncited by 322
- 2021ActionCLIP: A New Paradigm for Video Action Recognitioncited by 320
31 more
- 2023WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentationcited by 315
- 2023AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detectioncited by 271
- 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 200
- 2023APRIL-GAN: A Zero-/Few-Shot Anomaly Classification and Segmentation Method for CVPR 2023 VAND Workshop Challenge Tracks 1&2: 1st Place on Zero-shot AD and 4th Place on Few-shot ADcited by 130
- 2021A Hybrid Video Anomaly Detection Framework via Memory-Augmented Flow Reconstruction and Flow-Guided Frame Predictioncited by 129
- 2024AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detectioncited by 127
- 2019Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videoscited by 104
- 2019Graph Embedded Pose Clustering for Anomaly Detectioncited by 95
- 2021UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detectioncited by 93
- 2024Harnessing Large Language Models for Training-free Video Anomaly Detectioncited by 87
- 2022Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detectioncited by 72
- 2022Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzlescited by 62
- 2022Normalizing Flows for Human Pose Anomaly Detectioncited by 60
- 2025Towards Zero-Shot Anomaly Detection and Reasoning with Multimodal Large Language Modelscited by 54
- 2024FiLo: Zero-Shot Anomaly Detection by Fine-Grained Description and High-Quality Localizationcited by 51
- 2024ARC: A Generalist Graph Anomaly Detector with In-Context Learningcited by 48
- 2022SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detectioncited by 45
- 2022Generative Action Description Prompts for Skeleton-based Action Recognitioncited by 43
- 2023A New Comprehensive Benchmark for Semi-supervised Video Anomaly Detection and Anticipationcited by 42
- 2023Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detectioncited by 42
- 2023Hierarchical Semantic Contrast for Scene-aware Video Anomaly Detectioncited by 39
- 2024Generalized Out-of-Distribution Detection and Beyond in Vision Language Model Era: A Surveycited by 36
- 2025TSPO: Temporal Sampling Policy Optimization for Long-form Video Language Understandingcited by 25
- 2021Regularity Learning via Explicit Distribution Modeling for Skeletal Video Anomaly Detectioncited by 22
- 2024MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detectioncited by 22
- 2023Prompt-Guided Zero-Shot Anomaly Action Recognition using Pretrained Deep Skeleton Featurescited by 21
- 2022Cross-Domain Video Anomaly Detection without Target Domain Adaptationcited by 20
- 2024DA-Flow: Dual Attention Normalizing Flow for Skeleton-based Video Anomaly Detectioncited by 7
- 2024Frequency-Guided Diffusion Model with Perturbation Training for Skeleton-Based Video Anomaly Detectioncited by 6
- 2024Learning Discriminative Spatio-temporal Representations for Semi-supervised Action Recognitioncited by 2
- 2024Advancing Pre-trained Teacher: Towards Robust Feature Discrepancy for Anomaly Detectioncited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.