Benutzer: Gast  Login
Titel:

CertainNet: Sampling-free Uncertainty Estimation for Object Detection

Dokumenttyp:
Zeitschriftenaufsatz
Autor(en):
Stefano Gasperini; Jan Haug; Mohammad-Ali Nikouei Mahani; Alvaro Marcos-Ramiro; Nassir Navab; Benjamin Busam; Federico Tombari
Abstract:
Estimating the uncertainty of a neural network plays a fundamental role in safety-critical settings. In perception for autonomous driving, measuring the uncertainty means providing additional calibrated information to downstream tasks, such as path planning, that can use it towards safe navigation. In this work, we propose a novel sampling-free uncertainty estimation method for object detection. We call it CertainNet, and it is the first to provide separate uncertainties for each output signal:...     »
Stichworte:
uncertainty estimation; object detection; autonomous driving; deep learning; machine learning
Dewey Dezimalklassifikation:
000 Informatik, Wissen, Systeme
Zeitschriftentitel:
IEEE Robotics and Automation Letters (RA-L)
Jahr:
2021
Monat:
Nov
Reviewed:
ja
Sprache:
en
Volltext / DOI:
doi:10.1109/LRA.2021.3130976
Hinweise:
The first two authors contributed equally.
Status:
Verlagsversion / published
Copyright Informationen:
Copyright with IEEE.
 BibTeX