mlsanityData to 7 Oct 2026

UniTraj: A Unified Framework for Scalable Vehicle Trajectory Prediction

arXiv 2403.15098 ↗First posted 22 Mar 2024cs.CVcited by 51

Not on the map. Why?

Abstract (arXiv)

Vehicle trajectory prediction has increasingly relied on data-driven solutions, but their ability to scale to different data domains and the impact of larger dataset sizes on their generalization remain under-explored. While these questions can be studied by employing multiple datasets, it is challenging due to several discrepancies, e.g., in data formats, map resolution, and semantic annotation types. To address these challenges, we introduce UniTraj, a comprehensive framework that unifies various datasets, models, and evaluation criteria, presenting new opportunities for the vehicle trajectory prediction field. In particular, using UniTraj, we conduct extensive experiments and find that model performance significantly drops when transferred to other datasets. However, enlarging data size and diversity can substantially improve performance, leading to a new state-of-the-art result for the nuScenes dataset. We provide insights into dataset characteristics to explain these findings. The code can be found here: https://github.com/vita-epfl/UniTraj

Key facts. Cited by 51 papers in the map (data through 7 Oct 2026). Builds on 47 of its 64 references that are in the map. First posted on arXiv 22 Mar 2024.

Builds on

47 of 64 references.

References · 47

  1. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  2. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  3. 2019Argoverse: 3D Tracking and Forecasting with Rich Mapscited by 728
  4. 2023Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecastingcited by 701
  5. 2021Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion Datasetcited by 532
  6. 2021NuPlan: A closed-loop ML-based planning benchmark for autonomous vehiclescited by 485
  7. 2020Learning Lane Graph Representations for Motion Forecastingcited by 380
  8. 2021MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learningcited by 281
  9. 2022Wayformer: Motion Forecasting via Simple & Efficient Attention Networkscited by 279
  10. 2022Motion Transformer with Global Intention Localization and Local Movement Refinementcited by 267
  11. 2021MultiPath++: Efficient Information Fusion and Trajectory Aggregation for Behavior Predictioncited by 246
  12. 2020One Thousand and One Hours: Self-driving Motion Prediction Datasetcited by 239
35 more
  1. 2019INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Mapscited by 231
  2. 2023MotionDiffuser: Controllable Multi-Agent Motion Prediction using Diffusioncited by 210
  3. 2023MotionLM: Multi-Agent Motion Forecasting as Language Modelingcited by 154
  4. 2022DeepCore: A Comprehensive Library for Coreset Selection in Deep Learningcited by 146
  5. 2019The inD Dataset: A Drone Dataset of Naturalistic Road User Trajectories at German Intersectionscited by 137
  6. 2021Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Predictioncited by 131
  7. 2020Human Trajectory Forecasting in Crowds: A Deep Learning Perspectivecited by 128
  8. 2021Simple multi-dataset detectioncited by 126
  9. 2023Dataset Distillation: A Comprehensive Reviewcited by 126
  10. 2021Shifts: A Dataset of Real Distributional Shift Across Multiple Large-Scale Taskscited by 100
  11. 2023ScenarioNet: Open-Source Platform for Large-Scale Traffic Scenario Simulation and Modelingcited by 60
  12. 2021On Exposing the Challenging Long Tail in Future Prediction of Traffic Actorscited by 48
  13. 2022AdvDO: Realistic Adversarial Attacks for Trajectory Predictioncited by 48
  14. 2021Human Trajectory Prediction via Counterfactual Analysiscited by 42
  15. 2021Towards Robust and Adaptive Motion Forecasting: A Causal Representation Perspectivecited by 40
  16. 2021Vehicle trajectory prediction works, but not everywherecited by 39
  17. 2021Are socially-aware trajectory prediction models really socially-aware?cited by 32
  18. 2022SSL-Lanes: Self-Supervised Learning for Motion Forecasting in Autonomous Drivingcited by 31
  19. 2023FEND: A Future Enhanced Distribution-Aware Contrastive Learning Framework for Long-tail Trajectory Predictioncited by 31
  20. 2023trajdata: A Unified Interface to Multiple Human Trajectory Datasetscited by 25
  21. 2023Towards Motion Forecasting with Real-World Perception Inputs: Are End-to-End Approaches Competitive?cited by 22
  22. 2020OpenTraj: Assessing Prediction Complexity in Human Trajectories Datasetscited by 20
  23. 2021Injecting Knowledge in Data-driven Vehicle Trajectory Predictorscited by 18
  24. 2022Learning Semantic Segmentation from Multiple Datasets with Label Shiftscited by 14
  25. 2022Uncertainty estimation for Cross-dataset performance in Trajectory predictioncited by 13
  26. 2023Improving the Generalizability of Trajectory Prediction Models with Frenet-Based Domain Normalizationcited by 13
  27. 2022Motion Style Transfer: Modular Low-Rank Adaptation for Deep Motion Forecastingcited by 12
  28. 2023Toward Reliable Human Pose Forecasting with Uncertaintycited by 11
  29. 2022Multi-dataset Training of Transformers for Robust Action Recognitioncited by 10
  30. 2023Adv3D: Generating Safety-Critical 3D Objects through Closed-Loop Simulationcited by 10
  31. 2022Trajectory Forecasting from Detection with Uncertainty-Aware Motion Encodingcited by 8
  32. 2021Multi-dataset Pretraining: A Unified Model for Semantic Segmentationcited by 7
  33. 2022The Atlas Benchmark: an Automated Evaluation Framework for Human Motion Predictioncited by 7
  34. 2021Raising context awareness in motion forecastingcited by 6
  35. 2023How Does Traffic Environment Quantitatively Affect the Autonomous Driving Prediction?cited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 13; 2025: 24; 2026: 14.

51 papers.

Sort:

By area:

  1. 2025UniOcc: A Unified Benchmark for Occupancy Forecasting and Prediction in Autonomous Drivingcited by 27
  2. 2025GaussRender: Learning 3D Occupancy with Gaussian Renderingcited by 17
  3. 2025Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directionscited by 14
  4. 2024TAROT: Targeted Data Selection via Optimal Transportcited by 9
  5. 2025TrajDiff: End-to-end Autonomous Driving without Perception Annotationcited by 8
  6. 2024Adaptive Prediction Ensemble: Improving Out-of-Distribution Generalization of Motion Forecastingcited by 7
  7. 2025Continuous Locomotive Crowd Behavior Generationcited by 7
  8. 2026MAD: Motion Appearance Decoupling for efficient Driving World Modelscited by 7
  9. 2025Foresight in Motion: Reinforcing Trajectory Prediction with Reward Heuristicscited by 6
  10. 2025Generative Active Learning for Long-tail Trajectory Prediction via Controllable Diffusion Modelcited by 6
  11. 2024Valeo4Cast: A Modular Approach to End-to-End Forecastingcited by 5
  12. 2024Motion Forecasting via Model-Based Risk Minimizationcited by 4
38 more
  1. 2024Annealed Winner-Takes-All for Motion Forecastingcited by 4
  2. 2024LoRD: Adapting Differentiable Driving Policies to Distribution Shiftscited by 4
  3. 2025Unified Human Localization and Trajectory Prediction with Monocular Visioncited by 4
  4. 2025Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planningcited by 4
  5. 2025Uncertainty-Aware Trajectory Prediction via Rule-Regularized Heteroscedastic Deep Classificationcited by 3
  6. 2024Toward Unified Practices in Trajectory Prediction Research on Bird's-Eye-View Datasetscited by 2
  7. 2024Traffic and Safety Rule Compliance of Humans in Diverse Driving Situationscited by 2
  8. 2024DECODE: Domain-aware Continual Domain Expansion for Motion Predictioncited by 2
  9. 2024PPT: Pretraining with Pseudo-Labeled Trajectories for Motion Forecastingcited by 2
  10. 2025AccidentSim: Generating Vehicle Collision Videos with Physically Realistic Collision Trajectories from Real-World Accident Reportscited by 2
  11. 2025RetroMotion: Retrocausal Motion Forecasting Models are Instructablecited by 2
  12. 2025A Generalizable Physics-Enhanced State Space Model for Long-Term Dynamics Forecasting in Complex Environmentscited by 2
  13. 2025Goal-based Trajectory Prediction for improved Cross-Dataset Generalizationcited by 2
  14. 2025STEP: Structured Training and Evaluation Platform for benchmarking trajectory prediction modelscited by 2
  15. 2026123D: Unifying Multi-Modal Autonomous Driving Data at Scalecited by 2
  16. 2024A Multi-Loss Strategy for Vehicle Trajectory Prediction: Combining Off-Road, Diversity, and Directional Consistency Lossescited by 1
  17. 2024AI-Powered CPS-Enabled Vulnerable-User-Aware Urban Transportation Digital Twin: Methods and Applicationscited by 1
  18. 2025DISC: Dataset for Analyzing Driving Styles In Simulated Crashes for Mixed Autonomycited by 1
  19. 2025EP-Diffuser: An Efficient Diffusion Model for Traffic Scene Generation and Prediction via Polynomial Representationscited by 1
  20. 2025LMFormer: Lane based Motion Prediction Transformercited by 1
  21. 2026HetroD: A High-Fidelity Drone Dataset and Benchmark for Autonomous Driving in Heterogeneous Trafficcited by 1
  22. 2024Data-Driven Traffic Simulation for an Intersection in a Metropoliscited by 0
  23. 2025HiLO: High-Level Object Fusion for Autonomous Driving using Transformerscited by 0
  24. 2025BEVTraj: Map-Free End-to-End Trajectory Prediction in Bird's-Eye View with Deformable Attention and Sparse Goal Proposalscited by 0
  25. 2025Ensemble of Pre-Trained Models for Long-Tailed Trajectory Predictioncited by 0
  26. 2025What Did I Learn? Operational Competence Assessment for AI-Based Trajectory Plannerscited by 0
  27. 2025Enhanced Motion Forecasting with Plug-and-Play Multimodal Large Language Modelscited by 0
  28. 2025UniPlanner: A Unified Motion Planning Framework for Autonomous Vehicle Decision-Making Systems via Multi-Dataset Integrationcited by 0
  29. 2026COTTA: Context-Aware Transfer Adaptation for Trajectory Prediction in Autonomous Drivingcited by 0
  30. 2026SAIL: Scene-aware Adaptive Iterative Learning for Long-Tail Trajectory Prediction in Autonomous Vehiclescited by 0
  31. 2026Forecasting the Past: Gradient-Based Distribution Shift Detection in Trajectory Predictioncited by 0
  32. 2026CityTrajBench: A Unified Benchmark for City-Scale Vehicle Trajectory Generationcited by 0
  33. 2026Rethinking Training & Inference for Forecasting: Linking Winner-Take-All back to GMMscited by 0
  34. 2026LAMP: Lane-Aligned Motion Primitives for Feasible Trajectory Predictioncited by 0
  35. 2026Unveiling Transferability in Trajectory Prediction via Latent Scene Embeddingscited by 0
  36. 2026BLInD: Learning Driver Intent as a Distribution over Future Ego Trajectoriescited by 0
  37. 2026Kernel-Based Metrics Learning for Uncertain Opponent Vehicle Trajectory Prediction in Autonomous Racingcited by 0
  38. 2026HAP: A Hand-Driven Active Perception Framework for Egocentric Head Motion Predictioncited by 0

50 most cited of 51 All 51 (JSON)