InSyn: Modeling Complex Interactions for Pedestrian Trajectory Prediction

2025-07-16

AI Summary

Existing pedestrian trajectory prediction methods predominantly rely on relative position modeling to capture interactions, yet struggle to represent fine-grained behavioral patterns—such as grouping or collision avoidance—leading to degraded performance in dense scenes. To address this, we propose InSyn, a Transformer-based Interaction Synchronization Network that explicitly models diverse synchronized interaction patterns for the first time. We further introduce a direction-sensitive social force mechanism to enhance behavioral plausibility. To mitigate error accumulation and divergence in early-time-step predictions, we design the Seq-Start of Seq (SSOS) training strategy, which significantly improves temporal sequence stability. InSyn achieves end-to-end fusion of relative positional encoding and direction-aware modules. Evaluated on the ETH/UCY benchmark, InSyn outperforms state-of-the-art methods, achieving a 6.58% reduction in initial-step Average Displacement Error (ADE) under high-density conditions.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems

Abstract

Accurate pedestrian trajectory prediction is crucial for intelligent applications, yet it remains highly challenging due to the complexity of interactions among pedestrians. Previous methods have primarily relied on relative positions to model pedestrian interactions; however, they tend to overlook specific interaction patterns such as paired walking or conflicting behaviors, limiting the prediction accuracy in crowded scenarios. To address this issue, we propose InSyn (Interaction-Synchronization Network), a novel Transformer-based model that explicitly captures diverse interaction patterns (e.g., walking in sync or conflicting) while effectively modeling direction-sensitive social behaviors. Additionally, we introduce a training strategy termed Seq-Start of Seq (SSOS), designed to alleviate the common issue of initial-step divergence in numerical time-series prediction. Experiments on the ETH and UCY datasets demonstrate that our model outperforms recent baselines significantly, especially in high-density scenarios. Furthermore, the SSOS strategy proves effective in improving sequential prediction performance, reducing the initial-step prediction error by approximately 6.58%.

Problem

• Model complex pedestrian interactions for accurate trajectory prediction
• Address overlooked interaction patterns like paired walking or conflicts
• Reduce initial-step divergence in time-series trajectory prediction

Innovation

• Transformer-based model capturing diverse interaction patterns
• Direction-sensitive social behavior modeling
• Seq-Start of Seq training reducing initial error