Show simple item record

dc.contributor.author
Lungu, Iulia-Alexandra
dc.contributor.author
Hu, Yuhuang
dc.contributor.author
Liu, Shih-Chii
dc.date.accessioned
2020-09-07T08:24:09Z
dc.date.available
2020-05-30T03:17:38Z
dc.date.available
2020-06-05T07:58:06Z
dc.date.available
2020-06-05T07:58:49Z
dc.date.available
2020-06-05T09:50:59Z
dc.date.available
2020-09-07T08:24:09Z
dc.date.issued
2020
dc.identifier.isbn
978-1-7281-4922-6
en_US
dc.identifier.isbn
978-1-7281-4923-3
en_US
dc.identifier.other
10.1109/AICAS48895.2020.9073996
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/417491
dc.description.abstract
Few-shot learning, a rapidly evolving theme in deep learning research, aims to endow artificial intelligence with the same ability of humans to assimilate new information very quickly. Siamese networks have been used in this context to learn similarity between image pairs and quickly classify novel objects. This work proposes an improved architecture and a novel training method that increases a 1-shot 5-way classification accuracy on 5 entirely novel classes by around 5%, 19%, 18% and 13% respectively compared to vanilla Siamese networks when tested on Omniglot, Tiny-Imagenet, CIFAR100 as well as a custom dataset recorded with an event-driven camera. These networks, when run on a Jetson TX2 GPU can be executed within 108 ms on average.
en_US
dc.language.iso
en
en_US
dc.publisher
IEEE
en_US
dc.subject
Event camera
en_US
dc.subject
Machine learning
en_US
dc.subject
One-shot learning
en_US
dc.subject
Computer vision
en_US
dc.subject
Siamese networks
en_US
dc.title
Multi-Resolution Siamese Networks for One-Shot Learning
en_US
dc.type
Conference Paper
dc.date.published
2020-04-23
ethz.book.title
2020 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS)
en_US
ethz.pages.start
183
en_US
ethz.pages.end
187
en_US
ethz.event
2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS 2020) (virtual)
en_US
ethz.event.location
Genova, Italy
en_US
ethz.event.date
August 31 - September 2, 2020
en_US
ethz.notes
Conference postponed due to Corona virus (COVID-19). Due to the Corona virus (COVID-19) the conference was conducted virtually.
en_US
ethz.identifier.scopus
ethz.publication.place
Piscataway, NJ
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02533 - Institut für Neuroinformatik / Institute of Neuroinformatics
ethz.date.deposited
2020-05-30T03:18:00Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2020-09-07T08:24:20Z
ethz.rosetta.lastUpdated
2021-02-15T17:01:37Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Multi-Resolution%20Siamese%20Networks%20for%20One-Shot%20Learning&rft.date=2020&rft.spage=183&rft.epage=187&rft.au=Lungu,%20Iulia-Alexandra&Hu,%20Yuhuang&Liu,%20Shih-Chii&rft.isbn=978-1-7281-4922-6&978-1-7281-4923-3&rft.genre=proceeding&rft_id=info:doi/10.1109/AICAS48895.2020.9073996&rft.btitle=2020%202nd%20IEEE%20International%20Conference%20on%20Artificial%20Intelligence%20Circuits%20and%20Systems%20(AICAS)
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record