Research map / Video / Video segmentation and tracking
Dual Memory Units with Uncertainty Regulation for Weakly Supervised Video Anomaly Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 both normal and abnormal memory banks instead of using only the normal memory bank (ablation) | UR-DMU · UCF-Crime | AUC: 85.19 -> 86.97 | Helped |
| Add both normal and abnormal memory banks instead of using only the normal memory bank (ablation) | UR-DMU · XD-Violence | AP: 80.40 -> 81.66 | Helped |
| Add DMU to the GL-MHSA feature backbone (ablation) | UR-DMU · UCF-Crime | AUC: 82.89 -> 85.53 | Helped |
| Add DMU to the GL-MHSA feature backbone (ablation) | UR-DMU · XD-Violence | AP: 75.67 -> 79.80 | Helped |
| Add NUL to GL-MHSA + DMU (ablation) | UR-DMU · UCF-Crime | AUC: 85.53 -> 86.97 | Helped |
| Add NUL to GL-MHSA + DMU (ablation) | UR-DMU · XD-Violence | AP: 79.80 -> 81.66 | Helped |
| Add L_dis to L_dm + L_trip + L_kl (ablation) | UR-DMU · UCF-Crime | AUC: 86.60 -> 86.97 | Helped |
| Add L_dis to L_dm + L_trip + L_kl (ablation) | UR-DMU · XD-Violence | AP: 80.79 -> 81.66 | Helped |
| Set abnormal and normal memory bank counts to 60,60 rather than 20,20 (ablation) | UR-DMU · UCF-Crime | AUC: 86.71 -> 86.97 | Mixed |
| Set abnormal and normal memory bank counts to 60,60 rather than 20,20 (ablation) | UR-DMU · XD-Violence | AP: 81.13 -> 81.66 | Helped |
| Add L_trip to L_dm + L_kl + L_dis (ablation) | UR-DMU · UCF-Crime | AUC: 82.78 -> 86.97 | Helped |
| Add L_trip to L_dm + L_kl + L_dis (ablation) | UR-DMU · XD-Violence | AP: 78.16 -> 81.66 | Helped |
Closest papers by meaning
- Learning Memory-Guided Normality for Anomaly Detection (CVPR 2020 · similarity 0.90)
- Learning Event Completeness for Weakly Supervised Video Anomaly Detection (ICML 2025 · similarity 0.89)
- Joint Learning of General and Diverse Patterns with Mixture of Memory Experts for Weakly-Supervised Video Anomaly Detection (CVPR 2026 · similarity 0.89)
- Prompt-Enhanced Multiple Instance Learning for Weakly Supervised Video Anomaly Detection (CVPR 2024 · similarity 0.88)
- UCF-Crime-DVS: A Novel Event-Based Dataset for Video Anomaly Detection with Spiking Neural Networks (AAAI 2025 · similarity 0.87)
- VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection (AAAI 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.