mlsanityData to 14 Oct 2026

Hazy Pedestrian Trajectory Prediction via Physical Priors and Graph-Mamba

arXiv 2509.24020 ↗First posted 28 Sep 2025cs.CVcited by 0

On the map: State-space sequence models (S4, Mamba) → Video, motion and generative Mamba models

Abstract (arXiv)

To address the issues of physical information degradation and ineffective pedestrian interaction modeling in pedestrian trajectory prediction under hazy weather conditions, we propose a deep learning model that combines physical priors of atmospheric scattering with topological modeling of pedestrian relationships. Specifically, we first construct a differentiable atmospheric scattering model that decouples haze concentration from light degradation through a network with physical parameter estimation, enabling the learning of haze-mitigated feature representations. Second, we design an adaptive scanning state space model for feature extraction. Our adaptive Mamba variant achieves a 78% inference speed increase over native Mamba while preserving long-range dependency modeling. Finally, to efficiently model pedestrian relationships, we develop a heterogeneous graph attention network, using graph matrices to model multi-granularity interactions between pedestrians and groups, combined with a spatio-temporal fusion module to capture the collaborative evolution patterns of pedestrian movements. Furthermore, we constructed a new pedestrian trajectory prediction dataset based on ETH/UCY to evaluate the effectiveness of the proposed method. Experiments show that our method reduces the minADE / minFDE metrics by 37.2% and 41.5%, respectively, compared to the SOTA models in dense haze scenarios (visibility < 30m), providing a new modeling paradigm for reliable perception in intelligent transportation systems in adverse environments.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Video, motion and generative Mamba models, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 26 of its 53 references that are in the map. First posted on arXiv 28 Sep 2025.

Builds on

26 of 53 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM13 Mar

Other references · 25

  1. 2018Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networkscited by 690
  2. 2020VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representationcited by 530
  3. 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 325
  4. 2018SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraintscited by 298
  5. 2021DenseTNT: End-to-end Trajectory Prediction from Dense Goal Setscited by 295
  6. 2020Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Predictioncited by 218
  7. 2019Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories with GANscited by 79
  8. 2022SocialVAE: Human Trajectory Prediction using Timewise Latentscited by 67
  9. 2024TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecastingcited by 64
  10. 2024Mamba in Speech: Towards an Alternative to Self-Attentioncited by 60
  11. 2019StarNet: Targeted Computation for Object Detection in Point Cloudscited by 46
  12. 2024SingularTrajectory: Universal Trajectory Predictor Using Diffusion Modelcited by 35
13 more
  1. 2022FIFO: Learning Fog-invariant Features for Foggy Scene Segmentationcited by 31
  2. 2021Review of Pedestrian Trajectory Prediction Methods: Comparing Deep Learning and Knowledge-based Approachescited by 30
  3. 2023SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Predictioncited by 27
  4. 2023Calibrating Uncertainty for Semi-Supervised Crowd Countingcited by 22
  5. 2024Adapting to Length Shift: FlexiLength Network for Trajectory Predictioncited by 22
  6. 2024MART: MultiscAle Relational Transformer Networks for Multi-agent Trajectory Predictioncited by 21
  7. 2021Conditional Variational Autoencoder for Learned Image Reconstructioncited by 13
  8. 2024Recurrent Aligned Network for Generalized Pedestrian Trajectory Predictioncited by 9
  9. 2024Pedestrian Trajectory Prediction with Missing Data: Datasets, Imputation, and Benchmarkingcited by 8
  10. 2025Pedestrian Trajectory Prediction Based on Social Interactions Learning With Random Weightscited by 8
  11. 2018ADBSCAN: Adaptive Density-Based Spatial Clustering of Applications with Noise for Identifying Clusters with Varying Densitiescited by 4
  12. 2025ASTRA: A Scene-aware TRAnsformer-based model for trajectory predictioncited by 2
  13. 2024Pedestrian Motion Prediction Using Transformer-based Behavior Clustering and Data-Driven Reachability Analysiscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Social Graph Mamba: Forecasting Pedestrian Movements Based on Social Context
  2. 2025ViTE: Virtual Graph Trajectory Expert Router for Pedestrian Trajectory Prediction
  3. 2025HazeFlow: Revisit Haze Physical Model as ODE and Non-Homogeneous Haze Generation for Real-World Dehazing
  4. 2026Social-Mamba: Socially-Aware Trajectory Forecasting with State-Space Models
  5. 2026Unified Pedestrian Path Prediction Using Inverse Reinforcement Learning
  6. 2025SceneAware: Scene-Constrained Pedestrian Trajectory Prediction with LLM-Guided Walkability
  7. 2023GDTS: Goal-Guided Diffusion Model with Tree Sampling for Multi-Modal Pedestrian Trajectory Prediction
  8. 2024Joint Pedestrian Trajectory Prediction through Posterior Sampling