English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
 
 
DownloadE-Mail
 Local TagsRelease HistoryDetailsSummary
  VConv-DAE: Deep Volumetric Shape Learning Without Object Labels

Sharma, A., Grau, O., & Fritz, M. (2016). VConv-DAE: Deep Volumetric Shape Learning Without Object Labels. In G. Hua, & H. Jégou (Eds.), Computer Vision - ECCV 2016 Workshops (pp. 236-250). Berlin: Springer. doi:10.1007/978-3-319-49409-8_20.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/11858/00-001M-0000-002B-0642-6 Version Permalink: https://hdl.handle.net/11858/00-001M-0000-002C-3C6A-8
Genre: Conference Paper
Latex : {VConv}-{DAE}: {D}eep Volumetric Shape Learning Without Object Labels

Files

show Files

Locators

show

Creators

show
hide
 Creators:
Sharma, Abhishek1, Author           
Grau, Oliver2, Author
Fritz, Mario1, Author           
Affiliations:
1Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society, ou_1116547              
2External Organizations, ou_persistent22              

Content

show
hide
Free keywords: Computer Science, Computer Vision and Pattern Recognition, cs.CV,Computer Science, Graphics, cs.GR
 Abstract: With the advent of affordable depth sensors, 3D capture becomes more and more ubiquitous and already has made its way into commercial products. Yet, capturing the geometry or complete shapes of everyday objects using scanning devices (eg. Kinect) still comes with several challenges that result in noise or even incomplete shapes. Recent success in deep learning has shown how to learn complex shape distributions in a data-driven way from large scale 3D CAD Model collections and to utilize them for 3D processing on volumetric representations and thereby circumventing problems of topology and tessellation. Prior work has shown encouraging results on problems ranging from shape completion to recognition. We provide an analysis of such approaches and discover that training as well as the resulting representation are strongly and unnecessarily tied to the notion of object labels. Furthermore, deep learning research argues ~\cite{Vincent08} that learning representation with over-complete model are more prone to overfitting compared to the approach that learns from noisy data. Thus, we investigate a full convolutional volumetric denoising auto encoder that is trained in a unsupervised fashion. It outperforms prior work on recognition as well as more challenging tasks like denoising and shape completion. In addition, our approach is atleast two order of magnitude faster at test time and thus, provides a path to scaling up 3D deep learning.

Details

show
hide
Language(s): eng - English
 Dates: 2016-04-132016-09-09201620162016
 Publication Status: Issued
 Pages: 16 p.
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: BibTex Citekey: Sharma_arXiv2016
DOI: 10.1007/978-3-319-49409-8_20
 Degree: -

Event

show
hide
Title: Geometry Meets Deep Learning ECCV 2016 Workshop
Place of Event: Amsterdam, The Netherlands
Start-/End Date: 2016-10-09 - 2016-10-09

Legal Case

show

Project information

show

Source 1

show
hide
Title: Computer Vision - ECCV 2016 Workshops
  Abbreviation : ECCV-W 2016
  Other : ECCV 2016
  Other : ECCV 2016 W6 Geometry Meets Deep Learning
  Subtitle : Amsterdam, The Netherlands, October 8 -10 and 15–16, 2016; Proceedings, Part III
Source Genre: Proceedings
 Creator(s):
Hua, Gang1, Editor
Jégou, Hervé1, Editor
Affiliations:
1 External Organizations, ou_persistent22            
Publ. Info: Berlin : Springer
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 236 - 250 Identifier: ISBN: 978-3-319-49408-1

Source 2

show
hide
Title: Lecture Notes in Computer Science
  Abbreviation : LNCS
Source Genre: Series
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: 9915 Sequence Number: - Start / End Page: - Identifier: -