Research map / Graph learning / Knowledge and temporal graphs
Uncovering the Missing Pattern: Unified Framework Towards Trajectory Imputation and 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 unified imputation and prediction with GC-VRNN | GC-VRNN · Basketball-TIP, Football-TIP, and Vehicle-TIP | I-L2 and P-L2 | Helped |
| Add Static Topology GCL (ablation) | MS-GNN · Basketball-TIP r=3 ft. | I-L2: 7.16 -> 7.03 | Helped |
| Add Static Topology GCL (ablation) | MS-GNN · Basketball-TIP r=3 ft. | P-L2: 9.44 -> 7.50 | Helped |
| Add Dynamic Learnable GCL (ablation) | MS-GNN · Basketball-TIP r=3 ft. | I-L2: 7.24 -> 7.03 | Helped |
| Add Dynamic Learnable GCL (ablation) | MS-GNN · Basketball-TIP r=3 ft. | P-L2: 9.46 -> 7.50 | Helped |
| Add Edge Conditioned GCL (ablation) | MS-GNN · Basketball-TIP r=3 ft. | P-L2: 16.38 -> 7.50 | Helped |
| Add Temporal Decay (ablation) | GC-VRNN · Basketball-TIP r=3 ft. | I-L2: 7.25 -> 7.03 | Helped |
| Add Temporal Decay (ablation) | GC-VRNN · Basketball-TIP r=3 ft. | P-L2: 7.70 -> 7.50 | Helped |
| Jointly train imputation and prediction (ablation) | GC-VRNN · Basketball-TIP r=3 ft. | P-L2: 28.74 -> 7.50 | Helped |
| Jointly train imputation and prediction (ablation) | GC-VRNN · Basketball-TIP r=3 ft. | I-L2: 7.21 -> 7.03 | Helped |
| Share the RNN connection between imputation and prediction streams (ablation) | GC-VRNN · Basketball-TIP r=3 ft. | P-L2: 16.38 -> 7.50 | Helped |
Closest papers by meaning
- MS-TIP: Imputation Aware Pedestrian Trajectory Prediction (ICML 2024 · similarity 0.89)
- ProDiff: Prototype-Guided Diffusion for Minimal Information Trajectory Imputation (ICML 2025 · similarity 0.88)
- Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarking (NeurIPS 2024 · similarity 0.87)
- Motif-aware Graph Neural Networks for Networked Time Series Imputation (AAAI 2025 · similarity 0.86)
- Graph Structure Learning for Spatial-Temporal Imputation: Adapting to Node and Feature Scales (AAAI 2025 · similarity 0.85)
- Generative Human Trajectory Recovery via Embedding-Space Conditional Diffusion (ICML 2025 · similarity 0.85)
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.