Show simple item record

dc.contributor.author
Sakaridis, Christos
dc.contributor.author
Dai, Dengxin
dc.contributor.author
Van Gool, Luc
dc.date.accessioned
2020-03-30T06:13:29Z
dc.date.available
2020-03-30T06:13:29Z
dc.date.issued
2019
dc.identifier.isbn
978-1-7281-4803-8
en_US
dc.identifier.isbn
978-1-7281-4804-5
en_US
dc.identifier.other
10.1109/ICCV.2019.00747
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/407166
dc.identifier.doi
10.3929/ethz-b-000388200
dc.description.abstract
Most progress in semantic segmentation reports on daytime images taken under favorable illumination conditions. We instead address the problem of semantic segmentation of nighttime images and improve the state-of-the-art, by adapting daytime models to nighttime without using nighttime annotations. Moreover, we design a new evaluation framework to address the substantial uncertainty of semantics in nighttime images. Our central contributions are: 1) a curriculum framework to gradually adapt semantic segmentation models from day to night via labeled synthetic images and unlabeled real images, both for progressively darker times of day, which exploits cross-time-of-day correspondences for the real images to guide the inference of their labels; 2) a novel uncertainty-aware annotation and evaluation framework and metric for semantic segmentation, designed for adverse conditions and including image regions beyond human recognition capability in the evaluation in a principled fashion; 3) the Dark Zurich dataset, which comprises 2416 unlabeled nighttime and 2920 unlabeled twilight images with correspondences to their daytime counterparts plus a set of 151 nighttime images with fine pixel-level annotations created with our protocol, which serves as a first benchmark to perform our novel evaluation. Experiments show that our guided curriculum adaptation significantly outperforms state-of-the-art methods on real nighttime sets both for standard metrics and our uncertainty-aware metric. Furthermore, our uncertainty-aware evaluation reveals that selective invalidation of predictions can lead to better results on data with ambiguous content such as our nighttime benchmark and profit safety-oriented applications which involve invalid inputs.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
IEEE
en_US
dc.rights.uri
http://rightsstatements.org/page/InC-NC/1.0/
dc.title
Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation
en_US
dc.type
Conference Paper
dc.rights.license
In Copyright - Non-Commercial Use Permitted
dc.date.published
2020-02-27
ethz.book.title
2019 IEEE/CVF International Conference on Computer Vision (ICCV)
en_US
ethz.pages.start
7373
en_US
ethz.pages.end
7382
en_US
ethz.size
10 p. accepted version
en_US
ethz.version.deposit
acceptedVersion
en_US
ethz.event
IEEE International Conference on Computer Vision (ICCV 2019)
en_US
ethz.event.location
Seoul, South Korea
en_US
ethz.event.date
October 27 - November 2, 2019
en_US
ethz.notes
Poster presented on October 31, 2019
en_US
ethz.identifier.wos
ethz.identifier.scopus
ethz.publication.place
Piscataway, NJ
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02652 - Institut für Bildverarbeitung / Computer Vision Laboratory::03514 - Van Gool, Luc / Van Gool, Luc
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02652 - Institut für Bildverarbeitung / Computer Vision Laboratory::03514 - Van Gool, Luc / Van Gool, Luc
en_US
ethz.relation.isPartOf
10.3929/ethz-b-000488609
ethz.date.deposited
2020-01-07T13:26:09Z
ethz.source
FORM
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2020-06-16T12:58:02Z
ethz.rosetta.lastUpdated
2021-02-15T09:43:07Z
ethz.rosetta.exportRequired
true
ethz.rosetta.versionExported
true
dc.identifier.olduri
http://hdl.handle.net/20.500.11850/388200
dc.identifier.olduri
http://hdl.handle.net/20.500.11850/406644
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Guided%20Curriculum%20Model%20Adaptation%20and%20Uncertainty-Aware%20Evaluation%20for%20Semantic%20Nighttime%20Image%20Segmentation&rft.date=2019&rft.spage=7373&rft.epage=7382&rft.au=Sakaridis,%20Christos&Dai,%20Dengxin&Van%20Gool,%20Luc&rft.isbn=978-1-7281-4803-8&978-1-7281-4804-5&rft.genre=proceeding&rft_id=info:doi/10.1109/ICCV.2019.00747&rft.btitle=2019%20IEEE/CVF%20International%20Conference%20on%20Computer%20Vision%20(ICCV)
 Search print copy at ETH Library

Files in this item

Thumbnail

Publication type

Show simple item record