Deep Visual Odometry With Adaptive Memory

We propose a novel deep visual odometry (VO) method that considers global information by selecting memory and refining poses. Existing learning-based methods take the VO task as a pure tracking problem via recovering camera poses from image snippets, leading to severe error accumulation. Global info...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transactions on pattern analysis and machine intelligence 2022-02, Vol.44 (2), p.940-954
Hauptverfasser:	Xue, Fei, Wang, Xin, Wang, Junqiu, Zha, Hongbin
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms attention Cameras History memory Modules Pose estimation recurrent neural networks Simultaneous localization and mapping System effectiveness Task analysis Tracking Tracking problem Visibility Visual odometry
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!