Learning from demonstration in the wild

Feryal Behbahani, Kyriacos Shiarlis, Xi Chen, Vitaly Kurin, Sudhanshu Kasewa, Ciprian Stirbu, Joao Gomes, Supratik Paul, Frans A. Oliehoek, More Authors

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

6 Citations (Scopus)
35 Downloads (Pure)

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.

Original languageEnglish
Title of host publication2019 International Conference on Robotics and Automation, ICRA 2019
PublisherIEEE
Pages775-781
Number of pages7
ISBN (Electronic)978-1-5386-6027-0
ISBN (Print)978-1-5386-8176-3
DOIs
Publication statusPublished - 1 May 2019
Event2019 International Conference on Robotics and Automation, ICRA 2019 - Montreal, Canada
Duration: 20 May 201924 May 2019

Conference

Conference2019 International Conference on Robotics and Automation, ICRA 2019
CountryCanada
CityMontreal
Period20/05/1924/05/19

Fingerprint Dive into the research topics of 'Learning from demonstration in the wild'. Together they form a unique fingerprint.

Cite this