Research map / Graph learning / Knowledge and temporal graphs
Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks
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 |
|---|---|---|---|
| Combine graph attention with the Bicycle-GAN latent encoder (ablation) | GAT trajectory forecasting model · AVG | FDE: 1.07 -> 1.00 | Helped |
| Add graph attention to the BiGAN variant (ablation) | BiGAN trajectory forecasting model · AVG | FDE: 1.09 -> 1.00 | Helped |
| Add the latent scene encoder to GAT (ablation) | GAT trajectory forecasting model · AVG | ADE: 0.52 -> 0.48 | Helped |
| Use Social-BiGAT instead of Sophie | trajectory forecasting · AVG | FDE: 1.15 -> 1.00 | Helped |
| Reduce the number of generated samples K | Social-BiGAT · AVG | ADE: 0.476 -> 0.606 | Hurt |
| Add a latent scene encoder to reduce output variance | generative trajectory forecasting · AVG across five scenes | ADE: S-GAN-P: 0.558 -> 0.846; Sophie: 0.526 -> 0.712; Social-BiGAT: 0.476 -> 0.606 | Helped |
Closest papers by meaning
- Social-Transmotion: Promptable Human Trajectory Prediction (ICLR 2024 · similarity 0.89)
- A Unified Environmental Network for Pedestrian Trajectory Prediction (AAAI 2024 · similarity 0.88)
- SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents (AAAI 2024 · similarity 0.88)
- GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes (CVPR 2024 · similarity 0.87)
- SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.87)
- Higher-order Relational Reasoning for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.87)
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.