Research map / 3D vision / 3D generation
Argoverse: 3D Tracking and Forecasting With Rich Maps
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 map-based ground removal instead of plane-fitting (ablation) | Baseline tracker · Argoverse 3D Tracking test, 50 m | MOTA: 49.3 -> 50.0 | Helped |
| use map-based ground removal instead of plane-fitting (ablation) | Baseline tracker · Argoverse 3D Tracking test, 100 m | MOTP-I: 0.28 -> 0.25 | Helped |
| use map-based ground removal instead of plane-fitting (ablation) | Baseline tracker · Argoverse 3D Tracking test, 30 m | MOTP-O: 13.7 -> 15.3 | Hurt |
| add lane-direction orientation initialization (ablation) | Baseline tracker · Argoverse 3D Tracking test, 100 m | MOTP-O: 27.7 -> 13.3 | Helped |
| add social features to LSTM inputs (ablation) | LSTM · Argoverse Forecasting, K=1 | minADE: 2.15 -> 2.15 | No clear effect |
| add social features to LSTM inputs (ablation) | LSTM · Argoverse Forecasting, K=1 | minFDE: 4.97 -> 4.95 | No clear effect |
| prune nearest-neighbor predictions that leave the drivable area | NN · Argoverse Forecasting, K=6 | minFDE: 3.29 -> 3.19 | Helped |
| use driveable-area pruning | NN · Argoverse Forecasting, K=6 | DAC: 0.87 -> 0.94 | Helped |
| use lane-centerline coordinates and map prior | NN · Argoverse Forecasting, K=1 | minFDE: 7.88 -> 8.12 | Hurt |
| use lane-centerline coordinates and map prior | NN · Argoverse Forecasting, K=6 | DAC: 0.87 -> 0.96 | Helped |
| allow multiple predictions along each centerline rather than one (ablation) | NN+map(prior) · Argoverse Forecasting, K=6 | minFDE: 5.32 -> 4.02 | Helped |
| use an LSTM instead of nearest-neighbor regression with map prior | Map-prior forecasting · Argoverse Forecasting, K=6 | minFDE: 5.32 -> 4.19 | Helped |
Closest papers by meaning
- UnO: Unsupervised Occupancy Fields for Perception and Forecasting (CVPR 2024 · similarity 0.88)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.87)
- Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving (ICLR 2025 · similarity 0.87)
- The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset (NeurIPS 2026 · similarity 0.87)
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.86)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.86)
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.