Conference item icon

Conference item

Invariant information clustering for unsupervised image classification and segmentation

Abstract:

We present a novel clustering objective that learns a neural network classifier from scratch, given only unlabelled data samples. The model discovers clusters that accurately match semantic classes, achieving state-of-the-art results in eight unsupervised clustering benchmarks spanning image classification and segmentation. These include STL10, an unsupervised variant of ImageNet, and CIFAR10, where we significantly beat the accuracy of our closest competitors by 6.6 and 9.5 absolute percenta...

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

Actions


Access Document


Files:
Publisher copy:
10.1109/ICCV.2019.00996
Publication website:
https://ieeexplore.ieee.org/xpl/conhome/8972782/proceeding

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Oxford college:
New College
Role:
Author
Publisher:
IEEE Publisher's website
Host title:
2019 IEEE/CVF International Conference on Computer Vision (ICCV)
Pages:
9864-9873
Publication date:
2020-02-27
Acceptance date:
2019-07-22
Event title:
International Conference on Computer Vision (ICCV)
Event location:
Seoul, Korea
Event website:
http://iccv2019.thecvf.com
Event start date:
2019-10-27
Event end date:
2019-11-02
DOI:
EISSN:
2380-7504
EISBN:
978-1-7281-4803-8
Language:
English
Keywords:
Pubs id:
pubs:1060084
UUID:
uuid:b7abbed8-9223-45af-9764-67853b15d4a2
Local pid:
pubs:1060084
Source identifiers:
1060084
Deposit date:
2019-10-04

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