Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Dynamic and Static Context-aware LSTM for Multi-agent Motion Prediction
Chaofan Tao, Qinhong Jiang, Lixin Duan, Ping Luo
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 queue-based temporal context (ICM) to LSTM (ablation) | LSTM baseline · ETH-UCY (avg of 5 subsets) | ADE (m): 0.61 -> 0.44 | Improved |
| add social-aware context module (SCM) (ablation) | DSCMP IC · ETH-UCY (avg of 5 subsets) | ADE (m): 0.44 -> 0.42 | Improved |
| add scene-guided learnable latent variable (ablation) | DSCMP IC-SC · ETH-UCY (avg of 5 subsets) | ADE (m): 0.42 -> 0.41 | Improved |
| add scene-guided learnable latent variable (ablation) | DSCMP IC-SC · ETH-UCY (avg of 5 subsets) | FDE (m): 0.85 -> 0.80 | Improved |
| add queue-based temporal context (ICM) (ablation) | LSTM baseline · SDD (avg over 1-4 s) | ADE (pixels): 2.38 -> 2.00 | Improved |
| add social-aware context module (SCM) (ablation) | DSCMP IC · SDD (avg over 1-4 s) | ADE (pixels): 2.00 -> 1.98 | Improved |
| add scene-guided latent variable (ablation) | DSCMP IC-SC · SDD (avg over 1-4 s) | ADE (pixels): 1.98 -> 1.94 | Improved |
| add scene-guided latent variable (ablation) | DSCMP IC-SC · ETH-UCY (ZARA1) | ADE (m): 0.35 -> 0.33 | Improved |
| increase queue length from 2 to 4 (ablation) | DSCMP · ETH-UCY non-linear (ETH subset) | ADE (m): 0.73 -> 0.70 | Improved |
| increase queue length from 2 to 4 (ablation) | DSCMP · ETH-UCY non-linear (ZARA1) | ADE (m): 0.36 -> 0.39 | Hurt |
| increase queue length from 2 to 4 (ablation) | DSCMP · ETH-UCY non-linear (avg) | ADE (m): 0.50 -> 0.50 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Sophie — SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraints (CVPR 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Social LSTM: Human Trajectory Prediction in Crowded Spaces (CVPR 2016 · similarity 0.90)
- SCAN: A Spatial Context Attentive Network for Joint Multi-Agent Intent Prediction (AAAI 2021 · similarity 0.88)
- SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction (ECCV 2020 · similarity 0.88)
- GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes (CVPR 2024 · similarity 0.87)
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · similarity 0.86)
- Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning (CVPR 2024 · 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.
