Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Prediction

TL;DR

STAR introduces a novel attention-based framework for pedestrian trajectory prediction that models complex spatio-temporal interactions using Transformer mechanisms, achieving state-of-the-art results on multiple datasets.

Contribution

The paper presents STAR, a Transformer-only approach with a new graph convolution and external memory for improved trajectory prediction.

Findings

Achieves state-of-the-art performance on 5 datasets.

Effectively models complex spatio-temporal interactions.

Utilizes attention mechanisms exclusively for prediction.

Abstract

Understanding crowd motion dynamics is critical to real-world applications, e.g., surveillance systems and autonomous driving. This is challenging because it requires effectively modeling the socially aware crowd spatial interaction and complex temporal dependencies. We believe attention is the most important factor for trajectory prediction. In this paper, we present STAR, a Spatio-Temporal grAph tRansformer framework, which tackles trajectory prediction by only attention mechanisms. STAR models intra-graph crowd interaction by TGConv, a novel Transformer-based graph convolution mechanism. The inter-graph temporal dependencies are modeled by separate temporal Transformers. STAR captures complex spatio-temporal interactions by interleaving between spatial and temporal Transformers. To calibrate the temporal prediction for the long-lasting effect of disappeared pedestrians, we introduce a read-writable external memory module, consistently being updated by the temporal Transformer. We show that with only attention mechanism, STAR achieves state-of-the-art performance on 5 commonly used real-world pedestrian prediction datasets.