Title
Learning Depth With Very Sparse Supervision
Abstract
Motivated by the astonishing capabilities of natural intelligent agents and inspired by theories from psychology, this paper explores the idea that perception gets coupled to 3D properties of the world via interaction with the environment. Existing works for depth estimation require either massive amounts of annotated training data or some form of hard-coded geometrical constraint. This paper expl...
Year
DOI
Venue
2020
10.1109/LRA.2020.3009067
IEEE Robotics and Automation Letters
Keywords
DocType
Volume
Cameras,Optical imaging,Robot sensing systems,Estimation,Three-dimensional displays,Optical sensors
Journal
5
Issue
ISSN
Citations 
4
2377-3766
0
PageRank 
References 
Authors
0.34
0
3
Name
Order
Citations
PageRank
Antonio Loquercio1585.43
Alexey Dosovitskiy2179780.48
Davide Scaramuzza32704154.51