Logo des Repositoriums
 

Region-Based CNNs for Pedestrian Gender Recognition in Visual Surveillance Environments

Kurzanzeige
dc.contributor.authorYaghoubi, Ehsan
dc.contributor.authorAlirezazadeh, Pendar
dc.contributor.authorAssunção, Eduardo
dc.contributor.authorNeves, João C.
dc.contributor.authorProença, Hugo
dc.contributor.editorBrömme, Arslan
dc.contributor.editorBusch, Christoph
dc.contributor.editorDantcheva, Antitza
dc.contributor.editorRathgeb, Christian
dc.contributor.editorUhl, Andreas
dc.date.accessioned2020-09-15T13:01:25Z
dc.date.available2020-09-15T13:01:25Z
dc.date.issued2019
dc.description.abstractInferring soft biometric labels in totally uncontrolled outdoor environments, such as surveillance scenarios, remains a challenge due to the low resolution of data and its covariates that might seriously compromise performance (e.g., occlusions and subjects pose). In this kind of data, even state-of-the-art deep-learning frameworks (such as ResNet) working in a holistic way, attain relatively poor performance, which was the main motivation for the work described in this paper. In particular, having noticed the main effect of the subjects’ “pose” factor, in this paper we describe a method that uses the body keypoints to estimate the subjects pose and define a set of regions of interest (e.g., head, torso, and legs). This information is used to learn appropriate classification models, specialized in different poses/body parts, which contributes to solid improvements in performance. This conclusion is supported by the experiments we conducted in multiple real-world outdoor scenarios, using the data acquired from advertising panels placed in crowded urban environments.en
dc.identifier.isbn978-3-88579-690-9
dc.identifier.pissn1617-5468
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/34226
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofBIOSIG 2019 - Proceedings of the 18th International Conference of the Biometrics Special Interest Group
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-297
dc.subjectPedestrian attribute recognition
dc.subjectskeleton detection
dc.subjectpose estimation
dc.subjectsegmentation.
dc.titleRegion-Based CNNs for Pedestrian Gender Recognition in Visual Surveillance Environmentsen
dc.typeText/Conference Paper
gi.citation.endPage172
gi.citation.publisherPlaceBonn
gi.citation.startPage165
gi.conference.date18.-20. September 2019
gi.conference.locationDarmstadt, Germany
gi.conference.sessiontitleFurther Conference Contributions

Dateien

Originalbündel
1 - 1 von 1
Vorschaubild
Name:
BIOSIG_2019_paper_1_update.pdf
Größe:
2.91 MB
Format:
Adobe Portable Document Format
Herunterladen

Sammlungen

P296 - BIOSIG 2019 - Proceedings of the 18th International Conference of the Biometrics Special Interest Group