Research map / Video / Video segmentation and tracking
Real-World Anomaly Detection in Surveillance Videos
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 |
|---|---|---|---|
| use deep MIL ranking with video-level normal and anomalous labels instead of the dictionary-based approach | UCF-Crime anomaly detection · UCF-Crime test set | AUC: 65.51 -> 75.41 | Helped |
| use deep MIL ranking with video-level labels instead of the deep auto-encoder | UCF-Crime anomaly detection · UCF-Crime test set | AUC: 50.6 -> 75.41 | Helped |
| use deep MIL ranking with video-level labels instead of a binary classifier | UCF-Crime anomaly detection · UCF-Crime test set | AUC: 50.0 -> 75.41 | Helped |
| add sparsity and temporal smoothness constraints (ablation) | Deep MIL ranking anomaly detector · UCF-Crime test set | AUC: 74.44 -> 75.41 | Helped |
| use overlapping multi-scale temporal segments (ablation) | Deep MIL ranking anomaly detector · UCF-Crime test set | detection accuracy | No clear effect |
| increase network depth (ablation) | Deep MIL ranking anomaly detector · UCF-Crime test set | detection accuracy | No clear effect |
| use TCNN instead of C3D for anomalous activity recognition | UCF-Crime anomalous activity recognition · UCF-Crime activity recognition | Accuracy: 23.0 -> 28.4 | Helped |
Closest papers by meaning
- Open-Vocabulary Video Anomaly Detection (CVPR 2024 · similarity 0.89)
- Track Any Anomalous Object:A Granular Video Anomaly Detection Pipeline (CVPR 2025 · similarity 0.88)
- MoniTor: Exploiting Large Language Models with Instruction for Online Video Anomaly Detection (NeurIPS 2025 · similarity 0.88)
- Harnessing Large Language Models for Training-free Video Anomaly Detection (CVPR 2024 · similarity 0.88)
- Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity (CVPR 2025 · similarity 0.88)
- Advancing Video Anomaly Detection: A Concise Review and a New Dataset (NeurIPS 2024 · similarity 0.88)
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.