Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Multi-agent Trajectory Prediction with Fuzzy Query Attention
Nitin Kamra, Hao Zhu, Dweep Kumarbhai Trivedi, Ming Zhang, Yan Liu
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 velocity correction from LSTM intents and interactions to inertia-only model (ablation) | FQA · Collisions | RMSE: 0.519 -> 0.176 | Improved |
| add interaction module to inertia+intent model (ablation) | FQA · Collisions | RMSE: 0.236 -> 0.176 | Improved |
| add interaction module to inertia+intent model (ablation) | FQA · Charges | RMSE: 0.523 -> 0.409 | Improved |
| add interaction module to inertia+intent model (ablation) | FQA · ETH-UCY | RMSE: 0.549 -> 0.540 | Improved |
| use fuzzy decision making instead of plain FC layers with equal parameters (ablation) | FQA · Collisions | RMSE: 0.234 -> 0.176 | Improved |
| use fuzzy decision making instead of plain FC layers with equal parameters (ablation) | FQA · Charges | RMSE: 0.505 -> 0.409 | Improved |
| use fuzzy decision making instead of plain FC layers with equal parameters (ablation) | FQA · NBA | RMSE: 5.518 -> 5.449 | Improved |
| use distance-cutoff edges instead of fully-connected graph (ablation) | FQA · NGsim | RMSE: 5.071 -> 5.814 | Hurt |
| use distance-cutoff edges instead of fully-connected graph (ablation) | FQA · Collisions | RMSE: 0.176 -> 0.175 | No clear effect |
| use distance-cutoff edges instead of fully-connected graph | GSAGE · NGsim | RMSE: 5.582 -> 5.901 | Hurt |
| use distance-cutoff edges instead of fully-connected graph | GSAGE · Charges | RMSE: 0.522 -> 0.456 | Improved |
| add handcrafted fuzzy approach decision (ablation) | FQA · NGsim | RMSE: 5.071 -> 4.801 | Improved |
Closest papers by meaning
- M2I: From Factored Marginal Trajectory Prediction to Interactive Prediction (CVPR 2022 · similarity 0.87)
- FJMP: Factorized Joint Multi-Agent Motion Prediction Over Learned Directed Acyclic Interaction Graphs (CVPR 2023 · similarity 0.86)
- SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction (ECCV 2020 · similarity 0.86)
- DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024 · similarity 0.85)
- Density-Adaptive Model Based on Motif Matrix for Multi-Agent Trajectory Prediction (CVPR 2024 · similarity 0.84)
- MS-TIP: Imputation Aware Pedestrian Trajectory Prediction (ICML 2024 · 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.
