Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
162 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Improving Transferability for Cross-domain Trajectory Prediction via Neural Stochastic Differential Equation (2312.15906v1)

Published 26 Dec 2023 in cs.CV

Abstract: Multi-agent trajectory prediction is crucial for various practical applications, spurring the construction of many large-scale trajectory datasets, including vehicles and pedestrians. However, discrepancies exist among datasets due to external factors and data acquisition strategies. External factors include geographical differences and driving styles, while data acquisition strategies include data acquisition rate, history/prediction length, and detector/tracker error. Consequently, the proficient performance of models trained on large-scale datasets has limited transferability on other small-size datasets, bounding the utilization of existing large-scale datasets. To address this limitation, we propose a method based on continuous and stochastic representations of Neural Stochastic Differential Equations (NSDE) for alleviating discrepancies due to data acquisition strategy. We utilize the benefits of continuous representation for handling arbitrary time steps and the use of stochastic representation for handling detector/tracker errors. Additionally, we propose a dataset-specific diffusion network and its training framework to handle dataset-specific detection/tracking errors. The effectiveness of our method is validated against state-of-the-art trajectory prediction models on the popular benchmark datasets: nuScenes, Argoverse, Lyft, INTERACTION, and Waymo Open Motion Dataset (WOMD). Improvement in performance gain on various source and target dataset configurations shows the generalized competence of our approach in addressing cross-dataset discrepancies.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (50)
  1. Latent time neural ordinary differential equations. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, 6010–6018.
  2. A Set of Control Points Conditioned Pedestrian Trajectory Prediction. In AAAI Conference on Artificial Intelligence.
  3. Non-probability sampling network for stochastic human trajectory prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 6477–6487.
  4. nuscenes: A multimodal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 11621–11631.
  5. Estimating Treatment Effects from Irregular Time Series Observations with Hidden Confounders. arXiv preprint arXiv:2303.02320.
  6. Argoverse: 3d tracking and forecasting with rich maps. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 8748–8757.
  7. Neural Ordinary Differential Equations. In Bengio, S.; Wallach, H.; Larochelle, H.; Grauman, K.; Cesa-Bianchi, N.; and Garnett, R., eds., Advances in Neural Information Processing Systems, volume 31. Curran Associates, Inc.
  8. GRU-ODE-Bayes: Continuous modeling of sporadically-observed time series. Advances in neural information processing systems, 32.
  9. Large scale interactive motion forecasting for autonomous driving: The waymo open motion dataset. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9710–9719.
  10. Causal Intervention for Human Trajectory Prediction with Cross Attention Mechanism. In AAAI Conference on Artificial Intelligence.
  11. Uncertainty estimation for Cross-dataset performance in Trajectory prediction. ArXiv, abs/2205.07310.
  12. google-research. 2021. torchsde. https://github.com/google-research/torchsde.
  13. Long Short-Term Memory. Neural Computation, 9(8): 1735–1780.
  14. One thousand and one hours: Self-driving motion prediction dataset. In Conference on Robot Learning, 409–418. PMLR.
  15. Planning-oriented Autonomous Driving. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.
  16. Propagating State Uncertainty Through Trajectory Forecasting. In 2022 International Conference on Robotics and Automation (ICRA), 2351–2358.
  17. Neural controlled differential equations for irregular time series. Advances in Neural Information Processing Systems, 33: 6696–6707.
  18. SDE-Net: Equipping Deep Neural Networks with Uncertainty Estimates. In International Conference on Machine Learning, 5405–5415. PMLR.
  19. Human trajectory forecasting in crowds: A deep learning perspective. IEEE Transactions on Intelligent Transportation Systems, 23(7): 7386–7400.
  20. MUSE-VAE: Multi-Scale VAE for Environment-Aware Long Term Trajectory Prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2221–2230.
  21. GRIN: Generative Relation and Intention Network for Multi-agent Trajectory Prediction. In Beygelzimer, A.; Dauphin, Y.; Liang, P.; and Vaughan, J. W., eds., Advances in Neural Information Processing Systems.
  22. Scalable gradients for stochastic differential equations. In International Conference on Artificial Intelligence and Statistics, 3870–3882. PMLR.
  23. Temporal pyramid network for pedestrian trajectory prediction with multi-supervision. In Proceedings of the AAAI conference on artificial intelligence, volume 35, 2029–2037.
  24. Neural sde: Stabilizing neural ode networks with stochastic noise. arXiv preprint arXiv:1906.02355.
  25. Shifts: A Dataset of Real Distributional Shift Across Multiple Large-Scale Tasks. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2).
  26. Forecasting Human Trajectory from Scene History. In Oh, A. H.; Agarwal, A.; Belgrave, D.; and Cho, K., eds., Advances in Neural Information Processing Systems.
  27. Neural ODE Processes. In International Conference on Learning Representations.
  28. Identifying Reflected Images From Object Detector in Indoor Environment Utilizing Depth Information. IEEE Robotics and Automation Letters, 6(2): 635–642.
  29. Leveraging Future Relationship Reasoning for Vehicle Trajectory Prediction. In The Eleventh International Conference on Learning Representations.
  30. Vid-ode: Continuous-time video generation with neural ordinary differential equation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, 2412–2422.
  31. D-CODE: Discovering Closed-form ODEs from Observed Trajectories. In International Conference on Learning Representations.
  32. Latent ordinary differential equations for irregularly-sampled time series. Advances in neural information processing systems, 32.
  33. Probabilistic tracklet scoring and inpainting for multiple object tracking. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 14329–14339.
  34. Trajectron++: Dynamically-feasible trajectory forecasting with heterogeneous data. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XVIII 16, 683–700. Springer.
  35. Social interpretable tree for pedestrian trajectory prediction. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, 2235–2243.
  36. Collaborative Uncertainty in Multi-Agent Trajectory Forecasting. In Beygelzimer, A.; Dauphin, Y.; Liang, P.; and Vaughan, J. W., eds., Advances in Neural Information Processing Systems.
  37. Neural stochastic differential equations: Deep latent gaussian models in the diffusion limit. arXiv preprint arXiv:1905.09883.
  38. Attention is all you need. Advances in neural information processing systems, 30.
  39. ATPFL: Automatic Trajectory Prediction Model Design Under Federated Learning Framework. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 6563–6572.
  40. Transferable and adaptable driving behavior prediction. arXiv preprint arXiv:2202.05140.
  41. WSiP: Wave Superposition Inspired Pooling for Dynamic Interactions-Aware Trajectory Prediction. In AAAI Conference on Artificial Intelligence.
  42. Social ODE: Multi-agent Trajectory Forecasting with Neural Ordinary Differential Equations. In Avidan, S.; Brostow, G.; Cissé, M.; Farinella, G. M.; and Hassner, T., eds., Computer Vision – ECCV 2022, 217–233. Cham: Springer Nature Switzerland. ISBN 978-3-031-20047-2.
  43. Whose Track Is It Anyway? Improving Robustness to Tracking Errors with Affinity-based Trajectory Prediction. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 6563–6572.
  44. MTP: Multi-hypothesis Tracking and Prediction for Reduced Error Propagation. IV.
  45. MTP-GO: Graph-Based Probabilistic Multi-Agent Trajectory Prediction with Neural ODEs. IEEE Transactions on Intelligent Vehicles, 1–14.
  46. Multi-Stream Representation Learning for Pedestrian Trajectory Prediction. In AAAI Conference on Artificial Intelligence.
  47. Adaptive Trajectory Prediction via Transferable GNN. In 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 6510–6521.
  48. Improving the Generalizability of Trajectory Prediction Models with Frenét-Based Domain Normalization. 2023 IEEE International Conference on Robotics and Automation (ICRA), 11562–11568.
  49. INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps. arXiv:1910.03088 [cs, eess].
  50. HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 8823–8833.
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now