Research map / Image generation / Super-resolution
Inferring High-Resolution Traffic Accident Risk Maps Based on Satellite Imagery and GPS Trajectories
Songtao He, Mohammad Amin Sadeghi, Sanjay Chawla, Mohammad Alizadeh, Hari Balakrishnan, Samuel Madden
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 satellite imagery to road map inputs (ablation) | Road model · 4-city average, no history | AP: 15.51 -> 19.81 | Improved |
| add GPS histogram to road map inputs (ablation) | Road model · 4-city average, no history | AP: 15.51 -> 23.01 | Improved |
| add satellite imagery on top of road and GPS histogram (ablation) | Road+GPS-Hist · 4-city average, no history | AP: 23.01 -> 23.15 | No clear effect |
| add satellite imagery to road map inputs (ablation) | Road model · 4-city average, with history | AP: 35.86 -> 35.10 | Hurt |
| add GPS histogram to road map inputs (ablation) | Road model · 4-city average, with history | AP: 35.86 -> 38.33 | Improved |
| add satellite imagery on top of GPS features (ablation) | Road+GPS-Feat. · 4-city average, with history | AP: 38.28 -> 39.11 | Improved |
| add satellite imagery on top of GPS histogram (ablation) | Road+GPS-Hist · 4-city average, with history | AP: 38.33 -> 37.79 | Hurt |
| add satellite imagery to road map inputs in sparse Boston (ablation) | Road model · Boston, with history | AP: 23.16 -> 21.09 | Hurt |
| add satellite imagery to road map inputs in dense Los Angeles (ablation) | Road model · LA, with history | AP: 43.48 -> 44.77 | Improved |
| use deep model with multimodal inputs instead of KDE | Road+GPS-Feat.+Sat. · 4-city average, with history | AP: 34.24 -> 39.11 | Improved |
| use 5m model instead of 100m low-resolution upper bound | Road+GPS-Feat.+Sat. · 4-city average, with history | AP: 21.37 -> 39.11 | Improved |
| replace aggregated GPS features with Deep Set on raw trajectories (ablation) | Road+GPS model · 4-city dataset | accuracy | Hurt |
Closest papers by meaning
- TraffiDent: A Dataset for Understanding the Interplay Between Traffic Dynamics and Incidents (NeurIPS 2025 · similarity 0.84)
- Accident Anticipation via Temporal Occurrence Prediction (NeurIPS 2025 · similarity 0.83)
- Progressive Risk Estimation for Accident Anticipation (NeurIPS 2026 · similarity 0.82)
- Uncertainty-Instructed Structure Injection for Generalizable HD Map Construction (CVPR 2025 · similarity 0.82)
- FT-AED: Benchmark Dataset for Early Freeway Traffic Anomalous Event Detection (NeurIPS 2024 · similarity 0.81)
- SafeMap: Robust HD Map Construction from Incomplete Observations (ICML 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.
