Hazy Pedestrian Trajectory Prediction via Physical Priors and Graph-Mamba

2025-09-28

AI Summary

To address the degradation of physical cues and weakened interaction modeling in pedestrian trajectory prediction under haze conditions, this paper proposes an end-to-end framework integrating atmospheric scattering physics priors with heterogeneous graph-based relational modeling. Methodologically: (1) a differentiable atmospheric scattering model is constructed to decouple haze concentration from illumination degradation, enabling physics-driven feature restoration; (2) an adaptive Mamba variant is designed to enhance long-range spatiotemporal dependency learning; (3) a heterogeneous graph attention network explicitly captures multi-granularity pedestrian interactions and coordinated motion patterns. Evaluated on dense haze scenes with visibility <30 m, the method achieves 37.2% and 41.5% reductions in minADE and minFDE, respectively, over state-of-the-art approaches. Additionally, we release the first fog-specific trajectory prediction benchmark—ETH/UCY-Fog—a fog-augmented extension of ETH/UCY—to establish a new standard for robust behavior prediction under adverse weather.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Reasoning under Uncertainty: Relational Probabilistic Models
• Computer Vision: Diffusion Models for Vision

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

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.

Problem

• Predicting pedestrian trajectories in hazy weather conditions
• Addressing physical information degradation and interaction modeling
• Improving accuracy and speed for intelligent transportation systems

Innovation

• Differentiable atmospheric scattering model decouples haze concentration
• Adaptive Mamba variant increases inference speed by 78%
• Heterogeneous graph attention network models pedestrian multi-granularity interactions