Wojke, Nicolai und Bewley, Alex (2018) Deep Cosine Metric Learning for Person Re-Identification. In: Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017. IEEE. IEEE Winter Conference on Applications of Computer Vision (WACV), 2018-03-12 - 2018-03-14, Lake Tahoe, NV/CA. doi: 10.1109/WACV.2018.00087.
PDF
3MB |
Kurzfassung
Metric learning aims to construct an embedding where two extracted features corresponding to the same identity are likely to be closer than features from different identities. This paper presents a method for learning such a feature space where the cosine similarity is effectively optimized through a simple re-parametrization of the conventional softmax classification regime. At test time, the final classification layer can be stripped from the network to facilitate nearest neighbor queries on unseen individuals using the cosine similarity metric. This approach presents a simple alternative to direct metric learning objectives such as siamese networks that have required sophisticated pair or triplet sampling strategies in the past. The method is evaluated on two large-scale pedestrian re-identification datasets where competitive results are achieved overall. In particular, we achieve better generalization on the test set compared to a network trained with triplet loss.
elib-URL des Eintrags: | https://elib.dlr.de/116408/ | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Dokumentart: | Konferenzbeitrag (Vortrag) | ||||||||||||
Titel: | Deep Cosine Metric Learning for Person Re-Identification | ||||||||||||
Autoren: |
| ||||||||||||
Datum: | 2018 | ||||||||||||
Erschienen in: | Proceedings - 2017 IEEE Winter Conference on Applications of Computer Vision, WACV 2017 | ||||||||||||
Referierte Publikation: | Ja | ||||||||||||
Open Access: | Ja | ||||||||||||
Gold Open Access: | Nein | ||||||||||||
In SCOPUS: | Ja | ||||||||||||
In ISI Web of Science: | Nein | ||||||||||||
DOI: | 10.1109/WACV.2018.00087 | ||||||||||||
Verlag: | IEEE | ||||||||||||
Status: | veröffentlicht | ||||||||||||
Stichwörter: | Person Re-Identification, Metric Learning, Convolutional Neural Networks | ||||||||||||
Veranstaltungstitel: | IEEE Winter Conference on Applications of Computer Vision (WACV) | ||||||||||||
Veranstaltungsort: | Lake Tahoe, NV/CA | ||||||||||||
Veranstaltungsart: | internationale Konferenz | ||||||||||||
Veranstaltungsbeginn: | 12 März 2018 | ||||||||||||
Veranstaltungsende: | 14 März 2018 | ||||||||||||
Veranstalter : | IEEE | ||||||||||||
HGF - Forschungsbereich: | Luftfahrt, Raumfahrt und Verkehr | ||||||||||||
HGF - Programm: | Verkehr | ||||||||||||
HGF - Programmthema: | Verkehrsmanagement (alt) | ||||||||||||
DLR - Schwerpunkt: | Verkehr | ||||||||||||
DLR - Forschungsgebiet: | V VM - Verkehrsmanagement | ||||||||||||
DLR - Teilgebiet (Projekt, Vorhaben): | V - I.MoVe (alt) | ||||||||||||
Standort: | Berlin-Adlershof | ||||||||||||
Institute & Einrichtungen: | Institut für Verkehrssystemtechnik > Datenerfassung und Informationsgewinnung | ||||||||||||
Hinterlegt von: | Wojke, Nicolai | ||||||||||||
Hinterlegt am: | 19 Dez 2017 14:50 | ||||||||||||
Letzte Änderung: | 24 Apr 2024 20:20 |
Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags