Research map / 3D vision / Depth and stereo
Learning to Generate Maps from Trajectories
Sijie Ruan, Cheng Long, Jie Bao, Chunyang Li, Zisheng Yu, Ruiyuan Li, Yuxuan Liang, Tianfu He, Yu Zheng
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 transition-view features to spatial-view features (ablation) | T2RNet · TaxiBJ | F1: 0.3434 -> 0.3678 | Improved |
| add transition-view features to spatial-view features (ablation) | T2RNet · TaxiJN | F1: 0.2123 -> 0.2156 | Improved |
| concatenation-based skip fusion (UNet) instead of summation (LinkNet) | encoder-decoder segmentation · TaxiBJ | F1: 0.3325 -> 0.3439 | Improved |
| add auxiliary road region prediction task (multi-task T2RNet) over UNet | UNet · TaxiBJ | F1: 0.3439 -> 0.3678 | Improved |
| add auxiliary road region prediction task (multi-task T2RNet) over UNet | UNet · TaxiJN | F1: 0.2079 -> 0.2156 | Improved |
| multi-task loss weight lambda set to 0.2 versus smaller or larger (ablation) | T2RNet · TaxiBJ and TaxiJN | centerline F1 | Improved |
| add topology construction post-processing (ablation) | DeepMG · TaxiBJ map | map F1 | Improved |
| use Line feature instead of Point feature (ablation) | T2RNet · TaxiBJ vs TaxiJN | F1: TaxiBJ 0.3305 -> 0.3097; TaxiJN 0.2030 -> 0.2096 | Mixed |
| link penalty factor alpha 1.0 vs 3.0 (1.4 chosen) (ablation) | Map refinement · qualitative maps | map quality | Mixed |
Closest papers by meaning
- PLMTrajRec: A Scalable and Generalizable Trajectory Recovery Method with Pre-trained Language Models (NeurIPS 2025 · similarity 0.86)
- Translating Images to Road Network: A Sequence-to-Sequence Perspective (TPAMI 2025 · similarity 0.86)
- XTraj: A Coarse-to-Fine Autoregressive Framework for Transferable Trajectory Generation (NeurIPS 2026 · similarity 0.85)
- DeepRoadMapper: Extracting Road Topology From Aerial Images (ICCV 2017 · similarity 0.85)
- TrajMamba: An Efficient and Semantic-rich Vehicle Trajectory Pre-training Model (NeurIPS 2025 · similarity 0.84)
- GTG: Generalizable Trajectory Generation Model for Urban Mobility (AAAI 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.
