AVGCN: Trajectory Prediction using Graph Convolutional Networks Guided by Human Attention

Abstract

Pedestrian trajectory prediction is a critical yet challenging task, especially for crowded scenes. We suggest that introducing an attention mechanism to infer the importance of different neighbors is critical for accurate trajectory prediction in scenes with varying crowd size. In this work, we propose a novel method, AVGCN, for trajectory prediction utilizing graph convolutional networks (GCN) based on human attention (A denotes attention, V denotes visual field constraints). First, we train an attention network that estimates the importance of neighboring pedestrians, using gaze data collected as subjects perform a bird's eye view crowd navigation task. Then, we incorporate the learned attention weights modulated by constraints on the pedestrian's visual field into a trajectory prediction network that uses a GCN to aggregate information from neighbors efficiently. AVGCN also considers the stochastic nature of pedestrian trajectories by taking advantage of variational trajectory prediction. Our approach achieves state-of-the-art performance on several trajectory prediction benchmarks, and the lowest average prediction error over all considered benchmarks.

arXiv comment7 pages, 4 figures

References in corpus (7)

• Semi-Supervised Classification with Graph Convolutional Networks
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks
• Self-Consistent Trajectory Autoencoder: Hierarchical Reinforcement Learning with Trajectory Embeddings
• DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents
• CoMoGCN: Coherent Motion Aware Trajectory Prediction with Graph Representation
• Collaborative Motion Prediction via Neural Motion Message Passing