Performance Assessment of Convolutional Neural Networks for Semantic Image Segmentation

Show simple item record
dc.contributor.authorLeipnitz, Alexander
dc.contributor.authorStrutz, Tilo
dc.contributor.authorJokisch, Oliver
dc.contributor.editorSkala, Václav
dc.date.accessioned2019-10-22T09:09:06Z
dc.date.available2019-10-22T09:09:06Z
dc.date.issued2019
dc.description.abstract-translatedConvolutional neural networks are applied successfully for image classification and object detection. Recently, they have been adopted to semantic segmentation tasks and several new network architectures have been proposed. With respect to automotive applications, the Cityscapes dataset is often used as a benchmark. It is one of the biggest datasets in this field and consists of a training, a validation, and a test set. While training and validation allow the optimisation of these nets, the test dataset can be used to evaluate their performance. Our investigations have shown that while these networks perform well for images of the Cityscapes dataset, their segmentation quality significantly drops when applied to new data. It seems that they have limited generalisation abilities. In order to find out whether the image content itself or other image properties cause this effect, we have carried out systematic investigations with modified Cityscapes data. We have found that camera-dependent image properties like brightness, contrast, or saturation can significantly influence the segmentation quality. This papers presents the results of these tests including eight state-of-the-art CNNs. It can be concluded that the out-of-the-boxusage of CNNs in real-world environments is not recommended.en
dc.format9 s.cs
dc.format.mimetypeapplication/odt
dc.identifier.citationWSCG 2019: full papers proceedings: 27. International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, p. 27-35.en
dc.identifier.doihttps://doi.org/10.24132/CSRN.2019.2901.1.4
dc.identifier.isbn978-80-86943-37-4 (CD/-ROM)
dc.identifier.issn2464–4617 (print)
dc.identifier.issn2464-4625 (CD/DVD)
dc.identifier.urihttp://hdl.handle.net/11025/35606
dc.language.isoenen
dc.publisherVáclav Skala - UNION Agencycs
dc.rights© Václav Skala - UNION Agencycs
dc.rights.accessopenAccessen
dc.subjectkonvoluční neuronová síťcs
dc.subjectsémantická segmentacecs
dc.subjectgeneralizační schopnostics
dc.subject.translatedconvolutional neural networken
dc.subject.translatedsemantic segmentationen
dc.subject.translatedgeneralisation abilitiesen
dc.titlePerformance Assessment of Convolutional Neural Networks for Semantic Image Segmentationen
dc.typekonferenční příspěvekcs
dc.typeconferenceObjecten
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen

Files

Original bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
Leipnitz.pdf
Size:
845.78 KB
Format:
Adobe Portable Document Format
Description:
Plný text
Download
License bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

WSCG 2019: Full Papers Proceedings
 OPEN License Selector