Abstract
<h2>Abstract</h2> Trajectory Prediction under diverse patterns has attracted increasing attention in multiple real-world applications ranging from urban traffic analysis to human motion understanding, among which graph convolution network (GCN) is frequently adopted with its superior ability in modeling the complex trajectory interactions among multiple humans. In this work, we propose a python package by enhancing GCN with class label information of the trajectory, such that we can explicitly model not only human trajectories but also that of other road users such as vehicles. This is done by integrating a label-embedded graph with the existing graph structure in the standard graph convolution layer. The flexibility and the portability of the package also allow researchers to employ it under more general multi-class sequential prediction tasks.
Highlights
Multi-class trajectory prediction is challenging due to the needs to model different types of trajectory with diverse velocities and patterns
Our idea is developed on a spatial–temporal graph convolutional neural network (STGCNN) with a purposefully designed adjacency matrix to enhance the trajectory prediction with semantic meanings
We introduce the use of class labels in the geometric-based graph structure of the trajectories to be encoded
Summary
Multi-class trajectory prediction is challenging due to the needs to model different types of trajectory with diverse velocities and patterns. The trajectory of a car or that of a bike may have different impacts on the future movement of a pedestrian, as pedestrians tend to keep a larger distance with a car For such a multiclass situation, we model the trajectory of the road user to be predicted as an undirected spatial graph. We introduce the use of class labels in the geometric-based graph structure of the trajectories to be encoded. This way, the predicted trajectory will be informed by the observed trajectories of nearby roadusers, and their categorical properties. The implemented software has formed a part in demonstrating superior prediction accuracy for crowd trajectories [1]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
