University of Hertfordshire

By the same authors

Deep Learning for Semantic Segmentation on Minimal Hardware

Research output: Contribution to conferencePaper

Links

View graph of relations
Original languageEnglish
Publication statusPublished - 15 Jul 2018
EventRoboCup 2018 Symposium - Palais des congrès, Montreal, Canada
Duration: 22 Jun 201822 Jun 2018
http://www.robocup2018.org/?page=symposium&lang=en

Conference

ConferenceRoboCup 2018 Symposium
CountryCanada
CityMontreal
Period22/06/1822/06/18
Internet address

Abstract

Deep learning has revolutionised many fields, but it is still challenging to transfer its success to small mobile robots with minimal hardware. Specifically, some work has been done to this effect in the RoboCup humanoid football domain, but results that are performant and efficient and still generally applicable outside of this domain are lacking. We propose an approach conceptually different from those taken previously. It is based on semantic segmentation and does achieve these desired properties. In detail, it is being able to process full VGA images in real-time on a low-power mobile processor. It can further handle multiple image dimensions without retraining, it does not require specific domain knowledge for achieving a high frame rate and it is applicable on a minimal mobile hardware.

Notes

12 pages, 5 figures, RoboCup International Symposium 2018

ID: 15818420