Stochastic trajectory prediction with social graph network

TL;DR

This paper introduces a novel social graph-based neural network for pedestrian trajectory prediction that captures asymmetric social relationships and models uncertainty through a temporal stochastic approach, improving accuracy in crowded scenes.

Contribution

It proposes a directed social graph for dynamic social behavior modeling and a temporal stochastic method for uncertainty estimation, advancing trajectory prediction accuracy.

Findings

Effective in crowded scenes

Outperforms existing methods

Captures asymmetric social relations

Abstract

Pedestrian trajectory prediction is a challenging task because of the complexity of real-world human social behaviors and uncertainty of the future motion. For the first issue, existing methods adopt fully connected topology for modeling the social behaviors, while ignoring non-symmetric pairwise relationships. To effectively capture social behaviors of relevant pedestrians, we utilize a directed social graph which is dynamically constructed on timely location and speed direction. Based on the social graph, we further propose a network to collect social effects and accumulate with individual representation, in order to generate destination-oriented and social-aware representations. For the second issue, instead of modeling the uncertainty of the entire future as a whole, we utilize a temporal stochastic method for sequentially learning a prior model of uncertainty during social interactions. The prediction on the next step is then generated by sampling on the prior model and progressively decoding with a hierarchical LSTMs. Experimental results on two public datasets show the effectiveness of our method, especially when predicting trajectories in very crowded scenes.