Depth from Camera Motion and Object Detection

This paper addresses the problem of learning to estimate the depth of detected objects given some measurement of camera motion (e.g., from robot kinematics or vehicle odometry). We achieve this by 1) designing a recurrent neural network (DBox) that estimates the depth of objects using a generalized...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2021-03
Hauptverfasser:	Griffin, Brent A, Corso, Jason J
Format:	Artikel
Sprache:	eng
Schlagworte:	Benchmarks Cameras Datasets End effectors Kinematics Motion perception Object recognition Odometers Perturbation Recurrent neural networks Robot dynamics Robotics
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!