Research map / Video / Video segmentation and tracking
A Discriminative Framework for Anomaly Detection in Large Videos
Allison Del Giorno, J. Andrew Bagnell, Martial Hebert
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 |
|---|---|---|---|
| add 10 shuffles (K=10) vs none, window 10, lambda=0.01 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8697 -> 0.8950 | Improved |
| add 10 shuffles, window 10, lambda=1 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8921 -> 0.8957 | Improved |
| add 10 shuffles, window 10, lambda=100 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8947 -> 0.8958 | Improved |
| add 10 shuffles, window 100, lambda=0.01 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8149 -> 0.8951 | Improved |
| add 10 shuffles, window 100, lambda=1 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8736 -> 0.8954 | Improved |
| add 10 shuffles, window 100, lambda=100 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8922 -> 0.8953 | Improved |
| add 10 shuffles, window 1000, lambda=0.01 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.7731 -> 0.8085 | Improved |
| increase window size from 10 to 1000 with K=10, lambda=1 (ablation) | permutation anomaly framework · Avenue (frame-based AUC) | AUC: 0.8957 -> 0.8085 | Hurt |
| increase regularization lambda from 0.01 to 100 without shuffles, window 100 (ablation) | permutation anomaly framework (K=0) · Avenue (frame-based AUC) | AUC: 0.8149 -> 0.8922 | Improved |
| increase regularization lambda from 0.01 to 100 with 10 shuffles, window 10 (ablation) | permutation anomaly framework (K=10) · Avenue (frame-based AUC) | AUC: 0.8950 -> 0.8958 | No clear effect |
| average scores over shuffles instead of median or maximum (ablation) | permutation anomaly framework · Anomaly score aggregation | AUC | Improved |
Closest papers by meaning
- Learning Memory-Guided Normality for Anomaly Detection (CVPR 2020 · similarity 0.89)
- MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection (CVPR 2024 · similarity 0.89)
- Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity (CVPR 2025 · similarity 0.89)
- VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree (NeurIPS 2025 · similarity 0.88)
- Open-Vocabulary Video Anomaly Detection (CVPR 2024 · similarity 0.88)
- Harnessing Large Language Models for Training-free Video Anomaly Detection (CVPR 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.
