Research map / Probabilistic and scientific modeling / Molecules and proteins
Long-Term On-Board Prediction of People in Traffic Scenes Under Uncertainty
Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
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 Bayesian modeling of epistemic uncertainty to aleatoric uncertainty (ablation) | LSTM bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | MSE (pixels): 750 -> 618 | Improved |
| Add Bayesian modeling of epistemic uncertainty to aleatoric uncertainty (ablation) | LSTM bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | negative log-likelihood: 5.54 -> 4.13 | Improved |
| Add heteroscedastic aleatoric uncertainty modeling (ablation) | LSTM bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | negative log-likelihood: 7.77 -> 5.54 | Improved |
| Add heteroscedastic aleatoric uncertainty modeling (ablation) | LSTM bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | MSE (pixels): 650 -> 750 | Hurt |
| Add predicted future odometry through a second stream (ablation) | LSTM-Bayesian pedestrian predictor · Cityscapes bounding-box prediction (|B_p|=8) | MSE (pixels): 546 -> 505 | Improved |
| Add predicted future odometry through a second stream (ablation) | LSTM-Bayesian pedestrian predictor · Cityscapes bounding-box prediction (|B_p|=8) | negative log-likelihood: 3.97 -> 3.92 | Improved |
| Add visual input to the odometry predictor (ablation) | LSTM odometry predictor · Cityscapes odometry prediction (|O_p|=8) | speed MSE (m/sec): 0.056 -> 0.048 | Improved |
| Add visual input to the odometry predictor (ablation) | LSTM odometry predictor · Cityscapes odometry prediction (|O_p|=8) | angle MSE (degrees): 0.94 -> 0.88 | Improved |
| Use bounding-box prediction instead of center-only predictions (ablation) | LSTM-Bayesian pedestrian predictor · Cityscapes bounding-box center prediction | MSE: 695 -> 648 | Improved |
| Increase conditioning sequence length from 4 to 8 (ablation) | LSTM-Bayesian bounding-box predictor · Cityscapes bounding-box prediction | MSE (pixels): 647 -> 618 | Improved |
| Increase LSTM size from 128 to 512 units (ablation) | homoscedastic LSTM bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | MSE (pixels): 650 -> 705 | Hurt |
| Increase LSTM size from 128 to 512 units (ablation) | LSTM-Bayesian bounding-box predictor · Cityscapes bounding-box prediction (|B_p|=8) | MSE (pixels): 618 -> 619 | No clear effect |
Closest papers by meaning
- Leveraging Driver Field-of-View for Multimodal Ego-Trajectory Prediction (ICLR 2025 · similarity 0.87)
- Driving in the Occupancy World: Vision-Centric 4D Occupancy Forecasting and Planning via World Models for Autonomous Driving (AAAI 2025 · similarity 0.85)
- BAT: Behavior-Aware Human-Like Trajectory Prediction for Autonomous Driving (AAAI 2024 · similarity 0.85)
- Trajeglish: Traffic Modeling as Next-Token Prediction (ICLR 2024 · similarity 0.85)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.85)
- GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes (CVPR 2024 · similarity 0.85)
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.
