Research map / Video / Video segmentation and tracking
Deep Anomaly Discovery From Unlabeled Videos via Normality Advantage and Self-Paced Refinement
Guang Yu, Siqi Wang, Zhiping Cai, Xinwang Liu, Chuanfu Xu, Chengkun Wu
Check 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 |
|---|---|---|---|
| Add self-paced refinement to LBR (ablation) | LBR · UCSDped1, partial mode | frame-level AUROC: 79.7% -> 81.1% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · UCSDped2, partial mode | frame-level AUROC: 90.9% -> 95.7% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · Avenue, partial mode | frame-level AUROC: 90.4% -> 92.8% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · ShanghaiTech, partial mode | frame-level AUROC: 71.7% -> 72.1% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · UCSDped1, merge mode | frame-level AUROC: 80.1% -> 80.9% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · UCSDped2, merge mode | frame-level AUROC: 91.8% -> 97.2% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · Avenue, merge mode | frame-level AUROC: 89.5% -> 90.7% | Improved |
| Add self-paced refinement to LBR (ablation) | LBR · ShanghaiTech, merge mode | frame-level AUROC: 71.7% -> 72.6% | Improved |
| Add motion enhancement (ablation) | LBR-SPR · UCSDped1, partial and merge modes | frame-level AUROC: 81.1% -> 81.1%; 79.4% -> 80.9% | Mixed |
| Add motion enhancement (ablation) | LBR-SPR · UCSDped2, partial and merge modes | frame-level AUROC: 93.3% -> 95.7%; 97.0% -> 97.2% | Improved |
| Add motion enhancement (ablation) | LBR-SPR · Avenue, partial and merge modes | frame-level AUROC: 88.5% -> 92.8%; 89.7% -> 90.7% | Improved |
| Add motion enhancement (ablation) | LBR-SPR · ShanghaiTech, partial and merge modes | frame-level AUROC: 71.1% -> 72.1%; 71.9% -> 72.6% | Improved |
Closest papers by meaning
- FrameShield: Adversarially Robust Video Anomaly Detection (NeurIPS 2025 · similarity 0.88)
- Harnessing Large Language Models for Training-free Video Anomaly Detection (CVPR 2024 · similarity 0.87)
- Advancing Video Anomaly Detection: A Concise Review and a New Dataset (NeurIPS 2024 · similarity 0.87)
- VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree (NeurIPS 2025 · similarity 0.87)
- Dual Conditioned Motion Diffusion for Pose-Based Video Anomaly Detection (AAAI 2025 · similarity 0.86)
- Open-Vocabulary Video Anomaly Detection (CVPR 2024 · similarity 0.86)
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.
