Learning Structured Representations of Spatial and Interactive Dynamics for Trajectory Prediction in Crowded Scenes

Context plays a significant role in the generation of motion for dynamic agents in interactive environments. This work proposes a modular method that utilises a learned model of the environment for motion prediction. This modularity explicitly allows for unsupervised adaptation of trajectory predict...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-01
Main Authors: Davchev, Todor, Burke, Michael, Ramamoorthy, Subramanian
Format: Article
Language:English
Subjects:
Ablation
Decoupling
Environment models
Multiagent systems
Representations
Robot arms
Tracking problem
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Context plays a significant role in the generation of motion for dynamic agents in interactive environments. This work proposes a modular method that utilises a learned model of the environment for motion prediction. This modularity explicitly allows for unsupervised adaptation of trajectory prediction models to unseen environments and new tasks by relying on unlabelled image data only. We model both the spatial and dynamic aspects of a given environment alongside the per agent motions. This results in more informed motion prediction and allows for performance comparable to the state-of-the-art. We highlight the model's prediction capability using a benchmark pedestrian prediction problem and a robot manipulation task and show that we can transfer the predictor across these tasks in a completely unsupervised way. The proposed approach allows for robust and label efficient forward modelling, and relaxes the need for full model re-training in new environments.
ISSN:2331-8422
DOI:10.48550/arxiv.1911.13044