Research map / 3D vision / 3D generation
SiT Dataset: Socially Interactive Pedestrian Trajectory Dataset for Social Navigation Robots
Jongwook Bae, Jungho Kim, Junyong Yun, Changwon Kang, Jeongseon Choi, Chanhyeok Kim, Junho Lee, Jungwook Choi, Jun Won Choi
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 semantic map scene encoding | Y-Net · SiT trajectory prediction | ADE5: 1.527 -> 1.361 | Improved |
| add semantic map scene encoding | Y-Net · SiT trajectory prediction | ADE20: 0.836 -> 0.675 | Improved |
| add semantic map scene encoding | Y-Net · SiT trajectory prediction | FDE20: 1.878 -> 1.547 | Improved |
| add semantic map scene encoding | NSP-SFM · SiT trajectory prediction | ADE5: 1.346 -> 1.061 | Improved |
| add semantic map scene encoding | NSP-SFM · SiT trajectory prediction | FDE20: 1.087 -> 0.925 | Improved |
| fuse camera with LiDAR instead of camera only | 3D detectors · SiT 3D pedestrian detection | mAP: 0.244 -> 0.531 | Improved |
| use SECOND voxel backbone instead of PointPillars backbone | CenterPoint · SiT 3D pedestrian detection | mAP: 0.414 -> 0.518 | Improved |
| use SECOND voxel backbone instead of PointPillars backbone | TransFusion · SiT 3D pedestrian detection | mAP: 0.390 -> 0.531 | Improved |
| use SECOND voxel backbone instead of PointPillars backbone | FutureDet · SiT end-to-end motion prediction | mAPf: 0.037 -> 0.053 | Improved |
| use CenterPoint detector instead of PointPillars with AB3DMOT | AB3DMOT · SiT 3D pedestrian tracking | sAMOTA: 0.4110 -> 0.4841 | Improved |
| use CenterPoint tracker instead of AB3DMOT | CenterPoint detector · SiT 3D pedestrian tracking | sAMOTA / IDS: 0.4841 -> 0.6070 sAMOTA; 554 -> 1136 IDS | Mixed |
| use camera+LiDAR TransFusion-V instead of LiDAR CenterPoint-V | detector · SiT 3D pedestrian detection | AP(0.25): 0.397 -> 0.318 | Hurt |
Closest papers by meaning
- Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarking (NeurIPS 2024 · similarity 0.92)
- Social-Transmotion: Promptable Human Trajectory Prediction (ICLR 2024 · similarity 0.90)
- LocoVR: Multiuser Indoor Locomotion Dataset in Virtual Reality (ICLR 2025 · similarity 0.88)
- SocialCVAE: Predicting Pedestrian Trajectory via Interaction Conditioned Latents (AAAI 2024 · similarity 0.88)
- SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Prediction (CVPR 2024 · similarity 0.87)
- MS-TIP: Imputation Aware Pedestrian Trajectory Prediction (ICML 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.
