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

BlenderProc: Reducing the Reality Gap with Photorealistic Rendering

Denninger, Maximilian and Sundermeyer, Martin and Winkelbauer, Dominik and Olefir, Dmitry and Hodan, Tomas and Zidan, Youssef and Elbadrawy, Mohamad and Knauer, Markus and Katam, Harinandan and Lodhi, Ahsan (2020) BlenderProc: Reducing the Reality Gap with Photorealistic Rendering. In: International Conference on Robotics: Sciene and Systems, RSS 2020. Robotics: Science and Systems (RSS), 12.-16. Juli 2020, Virtuell.

[img] PDF


BlenderProc is an open-source and modular pipeline for rendering photorealistic images of procedurally generated 3D scenes which can be used for training data-hungry deep learning models. The presented results on the tasks of instance segmentation and surface normal estimation suggest that our photorealistic training images reduce the gap between the synthetic training and real test domains, compared to less realistic training images combined with domain randomization. BlenderProc can be used to train models for various computer vision tasks such as semantic segmentation or estimation of depth, optical flow, and object pose. By offering standard modules for parameterizing and sampling materials, objects, cameras and lights, BlenderProc can simulate various real-world scenarios and provide means to systematically investigate the essential factors for sim2real transfer.

Item URL in elib:https://elib.dlr.de/139317/
Document Type:Conference or Workshop Item (Other)
Title:BlenderProc: Reducing the Reality Gap with Photorealistic Rendering
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Denninger, MaximilianMaximilian.Denninger (at) dlr.dehttps://orcid.org/0000-0002-1557-2234
Sundermeyer, MartinMartin.Sundermeyer (at) dlr.dehttps://orcid.org/0000-0003-0587-9643
Winkelbauer, DominikDominik.Winkelbauer (at) dlr.dehttps://orcid.org/0000-0001-7443-1071
Olefir, DmitryDmitry.Olefir (at) dlr.dehttps://orcid.org/0000-0001-5244-9676
Hodan, Tomashodantom (at) cmp.felk.cvut.czUNSPECIFIED
Zidan, YoussefYoussef.Zidan (at) dlr.deUNSPECIFIED
Elbadrawy, MohamadMohamad.Elbadrawy (at) dlr.deUNSPECIFIED
Knauer, MarkusMarkus.Knauer (at) dlr.dehttps://orcid.org/0000-0001-8229-9410
Katam, HarinandanHarinandan.Katam (at) dlr.deUNSPECIFIED
Lodhi, AhsanAhsan.Lodhi (at) dlr.deUNSPECIFIED
Date:12 July 2020
Journal or Publication Title:International Conference on Robotics: Sciene and Systems, RSS 2020
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Keywords:BlenderProc, Simulation, Deep Learning, Neural Networks, Sim2Real
Event Title:Robotics: Science and Systems (RSS)
Event Location:Virtuell
Event Type:international Conference
Event Dates:12.-16. Juli 2020
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) > Perception and Cognition
Deposited By: Denninger, Maximilian
Deposited On:08 Dec 2020 14:50
Last Modified:08 Dec 2020 14:50

Repository Staff Only: item control page

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