Research map / Probabilistic and scientific modeling / Molecules and proteins
Social-DPF: Socially Acceptable Distribution Prediction of Futures
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 |
|---|---|---|---|
| replace Social STGCNN with Social-DPF | trajectory prediction on ETH&UCY · ETH&UCY AVG | ADE (m): 0.44 -> 0.43 | Helped |
| replace Social STGCNN with Social-DPF | trajectory prediction on ETH&UCY · ETH&UCY AVG | FDE (m): 0.75 -> 0.63 | Helped |
| use Social-DPF instead of the strongest listed baseline for collision avoidance | trajectory prediction · ETH&UCY | colliding persons per frame (%): 0.202 -> 0.030; 0.187 -> 0.012; 2.408 -> 0.512; 0.186 -> 0.039; 0.833 -> 0.114 | Helped |
| use both motion and output gates instead of only the motion gate (ablation) | Social-DPF · ETH&UCY AVG | ADE (m): 0.66 -> 0.43 | Helped |
| use both motion and output gates instead of only the output gate (ablation) | Social-DPF · ETH&UCY AVG | ADE (m): 0.93 -> 0.43 | Helped |
| add Lcollision1 to Lmode (ablation) | Social-DPF with both gates · ETH&UCY AVG | ADE (m): 0.51 -> 0.49 | Helped |
| add Lcollision2 to Lmode (ablation) | Social-DPF with both gates · ETH&UCY AVG | ADE (m): 0.51 -> 0.47 | Helped |
| add Lcollision1 to Lmode (ablation) | Social-DPF with both gates · ETH&UCY AVG | FDE (m): 0.70 -> 0.68 | Helped |
| add Lcollision2 to Lmode (ablation) | Social-DPF with both gates · ETH&UCY AVG | FDE (m): 0.70 -> 0.66 | Helped |
| test with 20 samples from the distributions instead of the maximum-weight component means (ablation) | full Social-DPF · ETH&UCY AVG | FDE (m): 0.98 -> 0.63 | Helped |
Closest papers by meaning
- SocialMOIF: Multi-Order Intention Fusion for Pedestrian Trajectory Prediction (CVPR 2025 · similarity 0.88)
- Social Physics Informed Diffusion Model for Crowd Simulation (AAAI 2024 · similarity 0.86)
- SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents (AAAI 2024 · similarity 0.85)
- Higher-order Relational Reasoning for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.85)
- SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.85)
- Resisting Stochastic Risks in Diffusion Planners with the Trajectory Aggregation Tree (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.