Research map / Video / Video segmentation and tracking
Robust Complex Behaviour Modeling at 90Hz
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 joint context refinement via correlation analysis (ablation) | BCD-based behaviour model · QMUL Junction temporal segmentation | segmentation accuracy: 92.50% -> 95.00% | Helped |
| add joint context refinement (ablation) | BCD-based behaviour model · QMUL Junction rare events | overall TPR: 81% -> 89% | Helped |
| add joint context refinement (ablation) | BCD-based behaviour model · QMUL Junction rare events | overall FPR: 0.4% -> 0.2% | Helped |
| add joint context refinement (ablation) | BCD-based behaviour model · MIT Traffic rare events | overall TPR: 34% -> 38% | Helped |
| add joint context refinement (ablation) | BCD-based behaviour model · MIT Traffic rare events | overall FPR: 0.2% -> 0% | Helped |
| add joint context refinement (ablation) | BCD-based behaviour model · Subway rare events | overall TPR: 100% -> 100% | No clear effect |
| use proposed model instead of MCTM | rare event detection · QMUL Junction | overall FPR: 1.0% -> 0.2% | Helped |
| use proposed model instead of MCTM | rare event detection · QMUL Junction | overall TPR: 52% -> 89% | Helped |
| use proposed model instead of MCTM | rare event detection · MIT Traffic | overall TPR: 27% -> 38% | Helped |
| use proposed model instead of MCTM | rare event detection · Subway | overall TPR: 100% -> 100% | No clear effect |
| use proposed model instead of MCTM | online detection · speed | detection speed: 9.18Hz -> 91.47Hz | Helped |
| use proposed model instead of Lu's sparse coding method | online detection · speed | detection speed: 91.47Hz -> 227.37Hz | Hurt |
Closest papers by meaning
- Hidden Follower Detection: How Is the Gaze-Spacing Pattern Embodied in Frequency Domain? (AAAI 2024 · similarity 0.82)
- Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection (NeurIPS 2025 · similarity 0.82)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.82)
- Learning Extremely High Density Crowds as Active Matters (CVPR 2025 · similarity 0.81)
- MoniTor: Exploiting Large Language Models with Instruction for Online Video Anomaly Detection (NeurIPS 2025 · similarity 0.81)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.81)
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.