A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification

A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification

Bibliographic Details
Main Author: Feydt, Austin Pack
Language:English
Published: Case Western Reserve University School of Graduate Studies / OhioLINK 2019
Subjects:
Computer Science
Robotics
Machine Learning
Computer Vision
Simulation
Simulation reality gap
point cloud
3-D point cloud
Deep Learning
Object Recognition
Online Access:http://rave.ohiolink.edu/etdc/view?acc_num=case1558355175360648
id ndltd-OhioLink-oai-etd.ohiolink.edu-case1558355175360648
record_format oai_dc
spelling ndltd-OhioLink-oai-etd.ohiolink.edu-case15583551753606482021-08-03T07:11:12Z A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification Feydt, Austin Pack Computer Science Robotics Machine Learning Computer Vision Simulation Simulation reality gap point cloud 3-D point cloud Deep Learning Object Recognition Computer vision tasks require collecting large volumes of data, which can be a time consuming effort. Automating the collection process with simulations speeds up the process, at the cost of the virtual data not closely matching the physical data. Building upon a previous attempt to bridge this gap, this thesis proposes three nuances to improve the correspondence between simulated and physical 3-D point clouds and depth images. First, the same CAD files used for simulated data acquisition are also used to 3-D print physical models used for physical data acquisition. Second, a new projection method is developed to make better use of all information provided by the depth camera. Finally, all projection parameters are unified to prevent the deep learning model from developing a dependence on intensity scaling. A convolutional neural network is trained on the simulated data and evaluated on the physical data to determine the model’s generalization ability. 2019-08-26 English text Case Western Reserve University School of Graduate Studies / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=case1558355175360648 http://rave.ohiolink.edu/etdc/view?acc_num=case1558355175360648 unrestricted This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.
collection NDLTD
language English
sources NDLTD
topic Computer Science
Robotics
Machine Learning
Computer Vision
Simulation
Simulation reality gap
point cloud
3-D point cloud
Deep Learning
Object Recognition
spellingShingle Computer Science
Robotics
Machine Learning
Computer Vision
Simulation
Simulation reality gap
point cloud
3-D point cloud
Deep Learning
Object Recognition
Feydt, Austin Pack
A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
author Feydt, Austin Pack
author_facet Feydt, Austin Pack
author_sort Feydt, Austin Pack
title A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
title_short A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
title_full A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
title_fullStr A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
title_full_unstemmed A Higher-Fidelity Approach to Bridging the Simulation-Reality Gap for 3-D Object Classification
title_sort higher-fidelity approach to bridging the simulation-reality gap for 3-d object classification
publisher Case Western Reserve University School of Graduate Studies / OhioLINK
publishDate 2019
url http://rave.ohiolink.edu/etdc/view?acc_num=case1558355175360648
work_keys_str_mv AT feydtaustinpack ahigherfidelityapproachtobridgingthesimulationrealitygapfor3dobjectclassification
AT feydtaustinpack higherfidelityapproachtobridgingthesimulationrealitygapfor3dobjectclassification
_version_ 1719455584081674240

Similar Items