Research map / Probabilistic and scientific modeling / Molecules and proteins
Joint Prediction for Kinematic Trajectories in Vehicle-Pedestrian-Mixed Scenes
Huikun Bi, Zhong Fang, Tianlu Mao, Zhaoqi Wang, Zhigang Deng
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 OBB-based joint position and orientation prediction for vehicles (ablation) | VPLSTM-P-20 · NGSIM vehicles (Tpred=12) | ADE: 7.95 -> 2.99 | Improved |
| add mixed social pooling to the OBB model (ablation) | VPLSTM-O-20 · NGSIM vehicles (Tpred=12) | ADE: 5.67 -> 2.99 | Improved |
| sample 20 trajectories instead of 1 (ablation) | VPLSTM-OP · NGSIM vehicles (Tpred=12) | ADE: 10.46 -> 2.99 | Improved |
| add OBB kinematic modelling for vehicles (ablation) | VPLSTM-P-20 · TJI vehicles (Tpred=12) | ADE: 89.54 -> 29.79 | Improved |
| separate LSTMs and OBB for vehicles vs SGAN-20VP-20 | VP-LSTM · TJI vehicles (Tpred=12) | ADE: 39.86 -> 29.79 | Improved |
| separate LSTMs and OBB for vehicles vs SGAN-20VP-20 (low density) | VP-LSTM · BJI vehicles (Tpred=12) | ADE: 24.65 -> 24.33 | No clear effect |
| separate pedestrian and vehicle modelling vs shared LSTM | VP-LSTM · BJI pedestrians (Tpred=12) | ADE: 20.33 -> 6.39 | Improved |
| predict orientation jointly instead of using last orientation | VP-LSTM · TJI vehicles orientation (Tpred=12) | ADEO: 38.65 -> 33.69 | Improved |
Closest papers by meaning
- MX-LSTM: Mixing Tracklets and Vislets to Jointly Forecast Trajectories and Head Poses (CVPR 2018 · similarity 0.87)
- CF-LSTM: Cascaded Feature-Based Long Short-Term Networks for Predicting Pedestrian Trajectory (AAAI 2020 · similarity 0.87)
- Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarking (NeurIPS 2024 · similarity 0.86)
- GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes (CVPR 2024 · similarity 0.86)
- A Unified Environmental Network for Pedestrian Trajectory Prediction (AAAI 2024 · similarity 0.84)
- TrajCLIP: Pedestrian trajectory prediction method using contrastive learning and idempotent networks (NeurIPS 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.
