Research map / Graph learning / Knowledge and temporal graphs
A Set of Control Points Conditioned Pedestrian Trajectory Prediction
Read the paperCheck 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 control-point-conditioned prediction and trajectory refinement | Graph-TERN · ETH/UCY AVG | ADE: 0.31 -> 0.24 | Helped |
| Use control-point-conditioned prediction and trajectory refinement | Graph-TERN · SDD (20 samples) | ADE: 9.03 -> 8.43 | Helped |
| Use control-point-conditioned prediction and trajectory refinement | Graph-TERN · Train Station | ADE (meter): 0.30 -> 0.27 | Helped |
| Use K=3 control points instead of K=1 (ablation) | Graph-TERN · ETH/UCY AVG | FDE: 0.503 -> 0.433 | Helped |
| Increase control points from K=3 to K=12 (ablation) | Graph-TERN · ETH/UCY AVG | FDE: 0.433 -> 0.561 | Hurt |
| Add a multi-relational GCN instead of omitting the GCN (ablation) | Graph-TERN control-point prediction · ETH/UCY AVG | FDE: 0.472 -> 0.433 | Helped |
| Add inverse-distance and inverse-displacement graph relations (ablation) | Graph-TERN control-point prediction · ETH/UCY AVG | FDE: 0.463 -> 0.433 | Helped |
| Add GMM pruning (ablation) | Graph-TERN control-point prediction · ETH/UCY AVG | FDE: 0.433 -> 0.385 | Helped |
| Add GMM pruning (ablation) | Graph-TERN trajectory refinement · ETH/UCY AVG | ADE: 0.255 -> 0.240 | Helped |
| Initialize refinement with endpoint interpolation from the last observed position rather than connecting control points (ablation) | Graph-TERN · ETH/UCY AVG | ADE: 0.348 -> 0.240 | Helped |
| Add guided endpoint sampling (ablation) | Graph-TERN refinement · ETH/UCY AVG | ADE: 0.246 -> 0.240 | Helped |
Closest papers by meaning
- A Unified Environmental Network for Pedestrian Trajectory Prediction (AAAI 2024 · similarity 0.90)
- Higher-order Relational Reasoning for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.90)
- TrajCLIP: Pedestrian trajectory prediction method using contrastive learning and idempotent networks (NeurIPS 2024 · similarity 0.88)
- Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarking (NeurIPS 2024 · similarity 0.88)
- MS-TIP: Imputation Aware Pedestrian Trajectory Prediction (ICML 2024 · similarity 0.88)
- SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents (AAAI 2024 · similarity 0.88)
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.