elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection

Sundermeyer, Martin and Marton, Zoltan-Csaba and Durner, Maximilian and Triebel, Rudolph (2020) Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection. International Journal of Computer Vision. Springer. doi: 10.1007/s11263-019-01243-8. ISSN 0920-5691.

[img] PDF - Only accessible within DLR - Published version
5MB

Abstract

We propose a real-time RGB-based pipeline for object detection and 6D pose estimation. Our novel 3D orientation estimation is based on a variant of the Denoising Autoencoder that is trained on simulated views of a 3D model using Domain Randomization. This so-called Augmented Autoencoder has several advantages over existing methods: It does not require real, pose-annotated training data, generalizes to various test sensors and inherently handles object and view symmetries. Instead of learning an explicit mapping from input images to object poses, it provides an implicit representation of object orientations defined by samples in a latent space. Our pipeline achieves state-of-the-art performance on the T-LESS dataset both in the RGB and RGB-D domain. We also evaluate on the LineMOD dataset where we can compete with other synthetically trained approaches. We further increase performance by correcting 3D orientation estimates to account for perspective errors when the object deviates from the image center and show extended results. Our code is available here https://github.com/DLR-RM/AugmentedAutoencoder.

Item URL in elib:https://elib.dlr.de/135549/
Document Type:Article
Title:Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Sundermeyer, Martinmartin.sundermeyer (at) dlr.dehttps://orcid.org/0000-0003-0587-9643
Marton, Zoltan-CsabaZoltan.Marton (at) dlr.dehttps://orcid.org/0000-0002-3035-493X
Durner, MaximilianMaximilian.Durner (at) dlr.dehttps://orcid.org/0000-0001-8885-5334
Triebel, RudolphRudolph.Triebel (at) dlr.dehttps://orcid.org/0000-0002-7975-036X
Date:March 2020
Journal or Publication Title:International Journal of Computer Vision
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:Yes
DOI :10.1007/s11263-019-01243-8
Publisher:Springer
ISSN:0920-5691
Status:Published
Keywords:6D Object Detection, Pose Estimation, Domain Randomization, Autoencoder, Synthetic Data, Pose Ambiguity, Symmetries
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Space System Technology
DLR - Research area:Raumfahrt
DLR - Program:R SY - Space System Technology
DLR - Research theme (Project):R - Vorhaben Multisensorielle Weltmodellierung (old)
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Robotics and Mechatronics (since 2013)
Deposited By: Sundermeyer, Martin
Deposited On:21 Jul 2020 09:47
Last Modified:23 Jul 2020 12:24

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.