Show simple item record

dc.contributor.authorMayuku, Orighomisan
dc.contributor.authorSurgenor, Brian W.
dc.contributor.authorMarshall, Joshua A.
dc.date.accessioned2021-07-12T15:56:39Z
dc.date.available2021-07-12T15:56:39Z
dc.date.issued2021-07-05
dc.identifier.urihttp://hdl.handle.net/1974/28953
dc.description.abstractFor use in off-road autonomous driving applications, we propose and study the use of multi-resolution local binary pat- tern texture descriptors to improve overall semantic segmentation performance and reduce class imbalance effects in off-road visual datasets. Our experiments, using a challenging publicly available off-road dataset as well as our own off-road dataset, show that texture features provide added flexibility towards reducing class imbalance effects, and that fusing color and texture features can improve segmentation performance. Finally, we demonstrate domain adaptation limitations in nominally similar off-road environments by cross-comparing the segmentation performance of convolutional neural networks trained on both datasets.en
dc.language.isoenen
dc.publisherIEEEen
dc.relationCollaborative Research & Development Grants Programen
dc.subjectimage featuresen
dc.subjectroboticsen
dc.subjectdomain adaptationen
dc.subjectimage segmentationen
dc.subjectcomputer visionen
dc.subjectautonomous vehiclesen
dc.titleMulti-resolution and multi-domain analysis of off-road datasets for autonomous drivingen
dc.typepreprinten
dc.identifier.doi10.1109/CRV52889.2021.00030
project.funder.identifierNSERC: http://dx.doi.org/10.13039/501100000038en
project.funder.nameNatural Sciences and Engineering Research Council of Canadaen
oaire.awardNumberDNDPJ 533392 - 18en


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record