Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Diverse and Admissible Trajectory Prediction through Multimodal Context Understanding
Seong Hyeon Park, Gyubok Lee, Jimin Seo, Manoj Bhat, Minseok Kang, Jonathan Francis, Ashwin Jadhav, Paul Pu Liang, Louis-Philippe Morency
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 cross-agent attention to the encoder (CAM vs CSP) | deterministic predictor · nuScenes | minADE: 1.390 -> 1.124 | Improved |
| add cross-agent attention to the encoder (CAM vs MATF-D) | deterministic predictor · nuScenes | minFDE: 2.538 -> 2.318 | Improved |
| add flow-based decoder to CAM (CAM-NF) | CAM · nuScenes | minADE: 1.124 -> 0.756 | Improved |
| add flow-based decoder to CAM (CAM-NF) | CAM · nuScenes | DAO: 3.121 -> 23.15 (Argoverse/nuScenes table: CAM 3.121, CAM-NF 23.15) | Improved |
| add agent-to-scene attention over the global scene (Global-CAM-NF) to CAM-NF (ablation) | CAM-NF · nuScenes | minFDE: 1.381 -> 1.357 | Improved |
| add local scene processing (Local-CAM-NF) to CAM-NF (ablation) | CAM-NF · nuScenes | minFDE: 1.381 -> 1.408 | No clear effect |
| add full agent-to-scene attention (AttGlobal) to Global-CAM-NF (ablation) | Global-CAM-NF · nuScenes | minFDE: 1.357 -> 1.171 | Improved |
| add full agent-to-scene attention (AttGlobal) to Global-CAM-NF (ablation) | Global-CAM-NF · Argoverse | minFDE: 1.252 -> 1.124 | Improved |
| add full agent-to-scene attention (AttGlobal) to Global-CAM-NF (ablation) | Global-CAM-NF · nuScenes | DAO: 22.65 -> 22.62 | Hurt |
| train with drivable-area-based annotation-free p instead of MSE (ablation) | AttGlobal-CAM-NF · nuScenes | minFDE: 1.379 -> 1.171 | Improved |
| train with drivable-area-based annotation-free p instead of MSE (ablation) | AttGlobal-CAM-NF · nuScenes | rF: 1.918 -> 2.558 | Improved |
| train with drivable-area-based annotation-free p instead of MSE (ablation) | AttGlobal-CAM-NF · nuScenes | DAC: 0.924 -> 0.918 | No clear effect |
Closest papers by meaning
- SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction (ECCV 2020 · similarity 0.93)
- ModeSeq: Taming Sparse Multimodal Motion Prediction with Sequential Mode Modeling (CVPR 2025 · similarity 0.89)
- DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024 · similarity 0.89)
- Generating Multimodal Driving Scenes via Next-Scene Prediction (CVPR 2025 · similarity 0.88)
- Motion Forecasting in Continuous Driving (NeurIPS 2024 · similarity 0.88)
- BehaviorGPT: Smart Agent Simulation for Autonomous Driving with Next-Patch Prediction (NeurIPS 2024 · similarity 0.87)
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.
