Deep-learning Approaches to Object Recognition from 3D Data

Deep-learning Approaches to Object Recognition from 3D Data

Bibliographic Details
Main Author: Chen, Zhiang
Language:English
Published: Case Western Reserve University School of Graduate Studies / OhioLINK 2017
Subjects:
Robotics
Computer Science
Nanoscience
Medical Imaging
deep learning
3D object recognition
semi-supervised learning
knowledge transfer
Online Access:http://rave.ohiolink.edu/etdc/view?acc_num=case1496303868914492
id ndltd-OhioLink-oai-etd.ohiolink.edu-case1496303868914492
record_format oai_dc
spelling ndltd-OhioLink-oai-etd.ohiolink.edu-case14963038689144922021-08-03T07:02:43Z Deep-learning Approaches to Object Recognition from 3D Data Chen, Zhiang Robotics Computer Science Nanoscience Medical Imaging deep learning 3D object recognition semi-supervised learning knowledge transfer This thesis focuses on deep-learning approaches to recognition and pose estimation of graspable objects using depth information. Recognition and orientation detection from depth-only data is encoded by a carefully designed 2D descriptor from 3D point clouds. Deep-learning approaches are explored from two main directions: supervised learning and semi-supervised learning. The disadvantages of supervised learning approaches drive the exploration of unsupervised pretraining. By learning good representations embedded in early layers, subsequent layers can be trained faster and with better performance. An understanding of learning processes from a probabilistic perspective is concluded, and it paves the way for developing networks based on Bayesian models, including Variational Auto-Encoders. Exploitation of knowledge transfer--re-using parameters learned from alternative training data--is shown to be effective in the present application. 2017-08-30 English text Case Western Reserve University School of Graduate Studies / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=case1496303868914492 http://rave.ohiolink.edu/etdc/view?acc_num=case1496303868914492 unrestricted This thesis or dissertation is protected by copyright: some rights reserved. It is licensed for use under a Creative Commons license. Specific terms and permissions are available from this document's record in the OhioLINK ETD Center.
collection NDLTD
language English
sources NDLTD
topic Robotics
Computer Science
Nanoscience
Medical Imaging
deep learning
3D object recognition
semi-supervised learning
knowledge transfer
spellingShingle Robotics
Computer Science
Nanoscience
Medical Imaging
deep learning
3D object recognition
semi-supervised learning
knowledge transfer
Chen, Zhiang
Deep-learning Approaches to Object Recognition from 3D Data
author Chen, Zhiang
author_facet Chen, Zhiang
author_sort Chen, Zhiang
title Deep-learning Approaches to Object Recognition from 3D Data
title_short Deep-learning Approaches to Object Recognition from 3D Data
title_full Deep-learning Approaches to Object Recognition from 3D Data
title_fullStr Deep-learning Approaches to Object Recognition from 3D Data
title_full_unstemmed Deep-learning Approaches to Object Recognition from 3D Data
title_sort deep-learning approaches to object recognition from 3d data
publisher Case Western Reserve University School of Graduate Studies / OhioLINK
publishDate 2017
url http://rave.ohiolink.edu/etdc/view?acc_num=case1496303868914492
work_keys_str_mv AT chenzhiang deeplearningapproachestoobjectrecognitionfrom3ddata
_version_ 1719452339151044608

Similar Items