Pedestrian Detector Domain Shift Robustness Evaluation, And Domain Shift Error Mitigation Proposal

Simple item page
but.event.date27.04.2021cs
but.event.titleSTUDENT EEICT 2021cs
dc.contributor.authorZemčík, Tomáš
dc.date.accessioned2023-01-06T10:05:44Z
dc.date.available2023-01-06T10:05:44Z
dc.date.issued2021cs
dc.description.abstractThis paper evaluates daytime to nighttime traffic image domain shift on Faster R-CNNand SSD based pedestrian and cyclist detectors. Daytime image trained detectors are applied on anewly compiled nighttime image dataset and their performance is evaluated against detectors trainedon both daytime and nighttime images. Faster R-CNN based detectors proved relatively robust, butstill clearly inferior to the models trained on nighttime images, the SSD based model proved noncompetitive.Approaches to the domain shift deterioration mitigation were proposed and future workoutlined.en
dc.formattextcs
dc.format.extent181-187cs
dc.format.mimetypeapplication/pdfen
dc.identifier.citationProceedings II of the 27st Conference STUDENT EEICT 2021: Selected Papers. s. 181-187. ISBN 978-80-214-5943-4cs
dc.identifier.doi10.13164/eeict.2021.181
dc.identifier.isbn978-80-214-5943-4
dc.identifier.urihttp://hdl.handle.net/11012/200837
dc.language.isoencs
dc.publisherVysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologiícs
dc.relation.ispartofProceedings II of the 27st Conference STUDENT EEICT 2021: Selected papersen
dc.relation.urihttps://conf.feec.vutbr.cz/eeict/index/pages/view/ke_stazenics
dc.rights© Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologiícs
dc.rights.accessopenAccessen
dc.subjectObject detectionen
dc.subjectPedestrian detectionen
dc.subjectCyclist detectionen
dc.subjectADASen
dc.subjectAVen
dc.subjectFaster R-CNN,SSDen
dc.subjectDomain shiften
dc.subjectDomain adaptationen
dc.subjectData augmentationen
dc.titlePedestrian Detector Domain Shift Robustness Evaluation, And Domain Shift Error Mitigation Proposalen
dc.type.driverconferenceObjecten
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.affiliatedInstitution.departmentFakulta elektrotechniky a komunikačních technologiícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
181_EEICT_2021_2.pdf
Size:
1.26 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Student EEICT 2021: Selected Papers