Geometric Graph Neural Network Modeling of Human Interactions in Crowded Environments
This work addresses the challenge of modeling pedestrian interactions for applications like autonomous navigation, but it is incremental as it builds on prior GNN methods by refining graph construction.
The paper tackled the problem of predicting human trajectories in crowded environments by proposing a geometric graph neural network that integrates psychological domain knowledge, resulting in improved prediction accuracy with reduced average and final displacement errors across multiple datasets.
Modeling human trajectories in crowded environments is challenging due to the complex nature of pedestrian behavior and interactions. This paper proposes a geometric graph neural network (GNN) architecture that integrates domain knowledge from psychological studies to model pedestrian interactions and predict future trajectories. Unlike prior studies using complete graphs, we define interaction neighborhoods using pedestrians' field of view, motion direction, and distance-based kernel functions to construct graph representations of crowds. Evaluations across multiple datasets demonstrate improved prediction accuracy through reduced average and final displacement error metrics. Our findings underscore the importance of integrating domain knowledge with data-driven approaches for effective modeling of human interactions in crowds.