Research map / Video / Video segmentation and tracking
Learning Anomalies with Normality Prior for Unsupervised Video Anomaly Detection
Haoyue Shi, Le Wang, Sanping Zhou, Gang Hua, Wei Tang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use normality prior (video start and end) as labeled normal snippets instead of random (ablation) | normality propagation · UCF-Crime | test AUC (%): 60.33 -> 75.99 | Improved |
| use normality prior instead of data-driven prior (ablation) | normality propagation · ShanghaiTech | test AUC (%): 83.59 -> 85.96 | Improved |
| use propagation with temporally-modulated feature similarity instead of direct comparison (ablation) | pseudo-label generation · ShanghaiTech | AUC (%): 72.33 -> 85.96 | Improved |
| combine temporal and feature similarity instead of either alone (ablation) | normality propagation · UCF-Crime | AUC (%): 63.01 -> 75.99 | Improved |
| add temporal-only similarity propagation versus direct comparison on UCF-Crime (ablation) | normality propagation · UCF-Crime | AUC (%): 71.22 -> 57.99 | Hurt |
| add loss re-weighting (ablation) | normality propagation · UCF-Crime | AUC (%): 75.99 -> 76.64 | Improved |
| vary number of high-confidence normal videos h from 0.1 to 0.5 (ablation) | loss re-weighting · ShanghaiTech and UCF-Crime | AUC (%): 86.46 -> 86.24 (ShanghaiTech) | No clear effect |
| use I3D features instead of ResNext with this method | proposed method · UCF-Crime | AUC (%): 80.02 vs C2FPL 80.65 | Mixed |
Closest papers by meaning
- Deep Anomaly Discovery From Unlabeled Videos via Normality Advantage and Self-Paced Refinement (CVPR 2022 · similarity 0.91)
- UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detection (CVPR 2022 · similarity 0.90)
- Advancing Video Anomaly Detection: A Concise Review and a New Dataset (NeurIPS 2024 · similarity 0.88)
- Open-Vocabulary Video Anomaly Detection (CVPR 2024 · similarity 0.88)
- UniVAD: A Training-free Unified Model for Few-shot Visual Anomaly Detection (CVPR 2025 · similarity 0.88)
- Harnessing Large Language Models for Training-free Video Anomaly Detection (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
