mlsanityData to 7 Oct 2026

Non-Probability Sampling Network for Stochastic Human Trajectory Prediction

arXiv 2203.13471 ↗First posted 25 Mar 2022cs.CV cs.LG cs.ROcited by 46

Not on the map. Why?

Abstract (arXiv)

Capturing multimodal natures is essential for stochastic pedestrian trajectory prediction, to infer a finite set of future trajectories. The inferred trajectories are based on observation paths and the latent vectors of potential decisions of pedestrians in the inference step. However, stochastic approaches provide varying results for the same data and parameter settings, due to the random sampling of the latent vector. In this paper, we analyze the problem by reconstructing and comparing probabilistic distributions from prediction samples and socially-acceptable paths, respectively. Through this analysis, we observe that the inferences of all stochastic models are biased toward the random sampling, and fail to generate a set of realistic paths from finite samples. The problem cannot be resolved unless an infinite number of samples is available, which is infeasible in practice. We introduce that the Quasi-Monte Carlo (QMC) method, ensuring uniform coverage on the sampling space, as an alternative to the conventional random sampling. With the same finite number of samples, the QMC improves all the multimodal prediction results. We take an additional step ahead by incorporating a learnable sampling network into the existing networks for trajectory prediction. For this purpose, we propose the Non-Probability Sampling Network (NPSN), a very small network (~5K parameters) that generates purposive sample sequences using the past paths of pedestrians and their social interactions. Extensive experiments confirm that NPSN can significantly improve both the prediction accuracy (up to 60%) and reliability of the public pedestrian trajectory prediction benchmark. Code is publicly available at https://github.com/inhwanbae/NPSN .

Key facts. Cited by 46 papers in the map (data through 7 Oct 2026). Builds on 31 of its 67 references that are in the map. First posted on arXiv 25 Mar 2022.

Builds on

31 of 67 references.

References · 31

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  5. 2018Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networkscited by 687
  6. 2020TNT: Target-driveN Trajectory Predictioncited by 345
  7. 2021AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecastingcited by 327
  8. 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 324
  9. 2017DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agentscited by 310
  10. 2018SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraintscited by 296
  11. 2020It Is Not the Journey but the Destination: Endpoint Conditioned Trajectory Predictioncited by 235
  12. 2019Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networkscited by 228
19 more
  1. 2020Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Predictioncited by 218
  2. 2017Social Attention: Modeling Attention in Human Crowdscited by 185
  3. 2019Multi-Agent Tensor Fusion for Contextual Trajectory Predictioncited by 131
  4. 2019Peeking into the Future: Predicting Future Person Activities and Locations in Videoscited by 112
  5. 2019SR-LSTM: State Refinement for LSTM towards Pedestrian Trajectory Predictioncited by 111
  6. 2021SGCN:Sparse Graph Convolution Network for Pedestrian Trajectory Predictioncited by 97
  7. 2016Learning in an Uncertain World: Representing Ambiguity Through Multiple Hypothesescited by 96
  8. 2018ChamNet: Towards Efficient Network Design through Platform-Aware Model Adaptationcited by 76
  9. 2019Conditional Generative Neural System for Probabilistic Trajectory Predictioncited by 76
  10. 2019The Garden of Forking Paths: Towards Multi-Future Trajectory Predictioncited by 72
  11. 2020MANTRA: Memory Augmented Networks for Multiple Trajectory Predictioncited by 69
  12. 2020Recursive Social Behavior Graph for Trajectory Predictioncited by 66
  13. 2021MG-GAN: A Multi-Generator Model Preventing Out-of-Distribution Samples in Pedestrian Trajectory Predictioncited by 61
  14. 2020Social NCE: Contrastive Learning of Socially-aware Motion Representationscited by 58
  15. 2021Human Trajectory Prediction via Counterfactual Analysiscited by 42
  16. 2020Reciprocal Learning Networks for Human Trajectory Predictioncited by 30
  17. 2020Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetrycited by 24
  18. 2020Dynamic and Static Context-aware LSTM for Multi-agent Motion Predictioncited by 21
  19. 2021A Robust Loss for Point Cloud Registrationcited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 13; 2024: 12; 2025: 10; 2026: 8.

46 papers.

Sort:

By area:

  1. 2023Leapfrog Diffusion Model for Stochastic Trajectory Predictioncited by 129
  2. 2023Leveraging Future Relationship Reasoning for Vehicle Trajectory Predictioncited by 47
  3. 2022Learning Pedestrian Group Representations for Multi-modal Trajectory Predictioncited by 44
  4. 2023EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecastingcited by 41
  5. 2024SingularTrajectory: Universal Trajectory Predictor Using Diffusion Modelcited by 35
  6. 2024Can Language Beat Numerical Regression? Language-Based Multimodal Trajectory Predictioncited by 31
  7. 2025MoFlow: One-Step Flow Matching for Human Trajectory Forecasting via Implicit Maximum Likelihood Estimation based Distillationcited by 31
  8. 2022Dynamic-Group-Aware Networks for Multi-Agent Trajectory Prediction with Relational Reasoningcited by 28
  9. 2024LG-Traj: LLM Guided Pedestrian Trajectory Predictioncited by 24
  10. 2024MART: MultiscAle Relational Transformer Networks for Multi-agent Trajectory Predictioncited by 21
  11. 2024Multi-Transmotion: Pre-trained Model for Human Motion Predictioncited by 21
  12. 2024T4P: Test-Time Training of Trajectory Prediction via Masked Autoencoder and Actor-specific Token Memorycited by 20
34 more
  1. 2023Unsupervised Sampling Promoting for Stochastic Human Trajectory Predictioncited by 17
  2. 2024Progressive Pretext Task Learning for Human Trajectory Predictioncited by 16
  3. 2025Unified Uncertainty-Aware Diffusion for Multi-Agent Trajectory Modelingcited by 15
  4. 2025Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directionscited by 14
  5. 2022Collaborative Uncertainty Benefits Multi-Agent Multi-Modal Trajectory Forecastingcited by 13
  6. 2023Improving Transferability for Cross-domain Trajectory Prediction via Neural Stochastic Differential Equationcited by 12
  7. 2024Intention-aware Denoising Diffusion Model for Trajectory Predictioncited by 11
  8. 2024MATRIX: Multi-Agent Trajectory Generation with Diverse Contextscited by 9
  9. 2023Vision-based Multi-future Trajectory Prediction: A Surveycited by 8
  10. 2023TrajPAC: Towards Robustness Verification of Pedestrian Trajectory Prediction Modelscited by 8
  11. 2023CRITERIA: a New Benchmarking Paradigm for Evaluating Trajectory Prediction Models for Autonomous Drivingcited by 8
  12. 2024Recurrent Aligned Network for Generalized Pedestrian Trajectory Predictioncited by 8
  13. 2025Continuous Locomotive Crowd Behavior Generationcited by 7
  14. 2025AMD: Adaptive Momentum and Decoupled Contrastive Learning Framework for Robust Long-Tail Trajectory Predictioncited by 4
  15. 2025ViTE: Virtual Graph Trajectory Expert Router for Pedestrian Trajectory Predictioncited by 4
  16. 2023Sparse Pedestrian Character Learning for Trajectory Predictioncited by 3
  17. 2024Evaluating Human Trajectory Prediction with Metamorphic Testingcited by 3
  18. 2025OmniTraj: Pre-Training on Heterogeneous Data for Adaptive and Zero-Shot Human Trajectory Predictioncited by 3
  19. 2024MetaTra: Meta-Learning for Generalized Trajectory Prediction in Unseen Domaincited by 2
  20. 2023Spatial-Temporal Conditional Random Field for Human Trajectory Predictioncited by 1
  21. 2025SceneAware: Scene-Constrained Pedestrian Trajectory Prediction with LLM-Guided Walkabilitycited by 1
  22. 2025VISTA: A Vision and Intent-Aware Social Attention Framework for Multi-Agent Trajectory Predictioncited by 1
  23. 2026ART: Adaptive Relational Transformer for Pedestrian Trajectory Prediction with Temporal-Aware Relationscited by 1
  24. 2023Visual Perception System for Autonomous Drivingcited by 0
  25. 2023Likely, Light, and Accurate Context-Free Clusters-based Trajectory Predictioncited by 0
  26. 2023Learning Socio-Temporal Graphs for Multi-Agent Trajectory Predictioncited by 0
  27. 2025IDRIFTNET: Physics-Driven Spatiotemporal Deep Learning for Iceberg Drift Forecastingcited by 0
  28. 2026SAIL: Scene-aware Adaptive Iterative Learning for Long-Tail Trajectory Prediction in Autonomous Vehiclescited by 0
  29. 2026Diverse Yet Consistent: Context-Guided Diffusion with Energy-Based Joint Refinement for Multi-Agent Motion Predictioncited by 0
  30. 2026Sparse Compositional Flow Matching by geometric assembly from motion primitivescited by 0
  31. 2026Agent-Centric Social Trajectory Prediction: A Free Energy Principle Perspectivecited by 0
  32. 2026TrajRS: Towards Certified Robustness in Pedestrian Trajectory Predictioncited by 0
  33. 2026Social Graph Mamba: Forecasting Pedestrian Movements Based on Social Contextcited by 0
  34. 2026Destination Support Restoration for Finite-Set Multimodal Trajectory Predictioncited by 0