Title
Learning from demonstration in the wild
Author
Behbahani, Feryal (Latent Logic)
Shiarlis, Kyriacos (Latent Logic)
Chen, Xi (Latent Logic)
Kurin, Vitaly (Latent Logic; University of Oxford)
Kasewa, Sudhanshu (Latent Logic; University of Oxford)
Stirbu, Ciprian (Latent Logic; University of Oxford)
Gomes, Joao (Latent Logic)
Paul, Supratik (Latent Logic; University of Oxford)
Oliehoek, F.A. (TU Delft Interactive Intelligence; Latent Logic)
Date
2019-05-01
Abstract
Learning from demonstration (LfD) is useful in settings where hand-coding behaviour or a reward function is impractical. It has succeeded in a wide range of problems but typically relies on manually generated demonstrations or specially deployed sensors and has not generally been able to leverage the copious demonstrations available in the wild: those that capture behaviours that were occurring anyway using sensors that were already deployed for another purpose, e.g., traffic camera footage capturing demonstrations of natural behaviour of vehicles, cyclists, and pedestrians. We propose video to behaviour (ViBe), a new approach to learn models of behaviour from unlabelled raw video data of a traffic scene collected from a single, monocular, initially uncalibrated camera with ordinary resolution. Our approach calibrates the camera, detects relevant objects, tracks them through time, and uses the resulting trajectories to perform LfD, yielding models of naturalistic behaviour. We apply ViBe to raw videos of a traffic intersection and show that it can learn purely from videos, without additional expert knowledge.
To reference this document use:
http://resolver.tudelft.nl/uuid:fed11b34-ef40-41d1-b7d7-5e2a3bf2127d
DOI
https://doi.org/10.1109/ICRA.2019.8794412
Publisher
IEEE
Embargo date
2020-02-12
ISBN
978-1-5386-8176-3
Source
2019 International Conference on Robotics and Automation, ICRA 2019
Event
2019 International Conference on Robotics and Automation, ICRA 2019, 2019-05-20 → 2019-05-24, Montreal, Canada
Bibliographical note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Part of collection
Institutional Repository
Document type
conference paper
Rights
© 2019 Feryal Behbahani, Kyriacos Shiarlis, Xi Chen, Vitaly Kurin, Sudhanshu Kasewa, Ciprian Stirbu, Joao Gomes, Supratik Paul, F.A. Oliehoek, More Authors