Conference item icon

Conference item

Salient deconvolutional networks

Abstract:

Deconvolution is a popular method for visualizing deep convolutional neural networks; however, due to their heuristic nature, the meaning of deconvolutional visualizations is not entirely clear. In this paper, we introduce a family of reversed networks that generalizes and relates deconvolution, backpropagation and network saliency. We use this construction to thoroughly investigate and compare these methods in terms of quality and meaning of the produced images, and of what architectural cho...

Expand abstract
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Files:
Publisher copy:
10.1007/978-3-319-46466-4_8

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Oxford college:
New College
Role:
Author
Publisher:
Springer Publisher's website
Volume:
9910
Pages:
120-135
Series:
Lecture Notes in Computer Science
Host title:
Computer Vision – ECCV 2016. ECCV 2016
Publication date:
2016-09-17
Acceptance date:
2016-07-11
Event title:
14th European Conference on Computer Vision (ECCV 2016)
Event location:
Amsterdam, The Netherlands
Event website:
http://www.eccv2016.org
Event start date:
2016-10-08T00:00:00Z
Event end date:
2016-10-16T00:00:00Z
DOI:
EISSN:
1611-3349
ISSN:
0302-9743
Source identifiers:
655292
ISBN:
9783319464657
Language:
English
Keywords:
Pubs id:
pubs:655292
UUID:
uuid:4a0a8851-1a3b-4af1-9fc7-0d796f9e09f5
Local pid:
pubs:655292
Deposit date:
2018-11-26

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP