Research map / Video / Video segmentation and tracking
Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videos
Check 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 global and local skeleton features with cross-branch message passing and combined reconstruction and prediction losses, rather than global and local branches without message passing (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.713 -> 0.754 | Helped |
| Add prediction loss alongside reconstruction loss (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.744 -> 0.754 | Helped |
| Add reconstruction loss alongside prediction loss (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.745 -> 0.754 | Helped |
| Use recurrent trajectory modeling instead of a frame-level image autoencoder (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.674 -> 0.754 | Helped |
| Use both global and local skeleton features with message passing instead of global features alone (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.689 -> 0.754 | Helped |
| Use both global and local skeleton features with message passing instead of local features alone (ablation) | MPED-RNN · HR-ShanghaiTech | frame-level ROC AUC: 0.715 -> 0.754 | Helped |
| Use the complete skeleton-based MPED-RNN instead of appearance-based Conv-AE | Video anomaly detection · HR-ShanghaiTech | frame-level ROC AUC: 0.698 -> 0.754 | Helped |
| Use the complete skeleton-based MPED-RNN instead of Liu et al.'s method | Video anomaly detection · HR-Avenue | frame-level ROC AUC: 0.862 -> 0.863 | Helped |
Closest papers by meaning
- Prompt-Guided Zero-Shot Anomaly Action Recognition Using Pretrained Deep Skeleton Features (CVPR 2023 · similarity 0.87)
- Dual Conditioned Motion Diffusion for Pose-Based Video Anomaly Detection (AAAI 2025 · similarity 0.87)
- Appearance-Motion Memory Consistency Network for Video Anomaly Detection (AAAI 2021 · similarity 0.87)
- Look Around for Anomalies: Weakly-Supervised Anomaly Detection via Context-Motion Relational Learning (CVPR 2023 · similarity 0.87)
- Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Model (AAAI 2025 · similarity 0.86)
- Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition (CVPR 2025 · similarity 0.84)
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.