STGT: Forecasting Pedestrian Motion Using Spatio-Temporal Graph Transformer
Document Type
Conference Proceeding
Publication Date
7-11-2021
Publication Title
IEEE Intelligent Vehicles Symposium, Proceedings
Publisher
IEEE Xplore
Publisher Location
Manhattan, New York
First page number:
1553
Last page number:
1558
Abstract
Full understanding human motion is essential for autonomous agents such as self-driving vehicles and social robots for navigating in dense crowded environments. In this paper, we present a trajectory prediction framework which models inter-pedestrian behaviour through graph representations and then apply attention through a Transformer network to better forecast human motion. Previous works have incorporated pedestrian interaction using social and graph pooling mechanisms whereas our work utilizes complete graph structure of pedestrians which helps to obtain robust spatiotemporal representations. We also leverage semantic segmentation architecture to encode scene context. Our experiments highlight the potential of handing pedestrian interaction with graph convolutional networks and Transformer and, on top of that, shows marginal improvement with inclusion of semantic scene features.
Controlled Subject
Pedestrians; Automated vehicles
Disciplines
Electrical and Computer Engineering | Systems and Communications
Repository Citation
Syed, A.,
Morris, B.
(2021).
STGT: Forecasting Pedestrian Motion Using Spatio-Temporal Graph Transformer.
IEEE Intelligent Vehicles Symposium, Proceedings
1553-1558.
Manhattan, New York: IEEE Xplore.
http://dx.doi.org/10.1109/IV48863.2021.9575498