Rethinking the Competition Between Detection and ReID in Multiobject Tracking

Due to balanced accuracy and speed, one-shot models which jointly learn detection and identification embeddings, have drawn great attention in multi-object tracking (MOT). However, the inherent differences and relations between detection and re-identification (ReID) are unconsciously overlooked beca...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on image processing Jg. 31; S. 3182 - 3196
Hauptverfasser:	Liang, Chao, Zhang, Zhipeng, Zhou, Xue, Li, Bing, Zhu, Shuyuan, Hu, Weiming
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	New York IEEE 2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:	Bells Competition Computational modeling Detectors Feature extraction ID embedding Misalignment Multiobject tracking Multiple target tracking Object detection one-shot reciprocal representation learning Representations scale-aware attention Semantics Target tracking Task analysis
ISSN:	1057-7149, 1941-0042, 1941-0042
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!