Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images

Hanna Ragnarsdóttir; Þórhildur Þorleiksdóttir; Omar Shahbaz Khan; Björn Thór Jónsson; Gylfi Þór Guðmundsson; Jan Zahálka; Stevan Rudinac; Laurent Amsaleg; Marcel Worring

doi:10.1145/3343031.3350580

Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images

Hanna Ragnarsdóttir, Þórhildur Þorleiksdóttir, Omar Shahbaz Khan, Björn Thór Jónsson, Gylfi Þór Guðmundsson, Jan Zahálka, Stevan Rudinac, Laurent Amsaleg, Marcel Worring

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

In this demonstration, we present Exquisitor, a media explorer capable of learning user preferences in real-time during interactions with the 99.2 million images of YFCC100M. Exquisitor owes its efficiency to innovations in data representation, compression, and indexing. Exquisitor can complete each interaction round, including learning preferences and presenting the most relevant results, in less than 30 ms using only a single CPU core and modest RAM. In short, Exquisitor can bring large-scale interactive learning to standard desktops and laptops, and even high-end mobile devices.

Original language	English
Title of host publication	Proceedings of the ACM Multimedia Conference
Number of pages	3
Place of Publication	Nice, France
Publisher	Association for Computing Machinery
Publication date	Oct 2019
Pages	1029-1031
ISBN (Electronic)	978-1-4503-6889-6
DOIs	https://doi.org/10.1145/3343031.3350580
Publication status	Published - Oct 2019

Keywords

Interactive multimodal learning
Scalability
100 million images

Access to Document

10.1145/3343031.3350580

Exquisitor__ACM_MM_demo_submissionSubmitted manuscript, 2.04 MB

https://dl.acm.org/citation.cfm?doid=3343031.3350580

Cite this

Ragnarsdóttir, H., Þorleiksdóttir, Þ., Khan, O. S., Jónsson, B. T., Guðmundsson, G. Þ., Zahálka, J., Rudinac, S., Amsaleg, L., & Worring, M. (2019). Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images. In Proceedings of the ACM Multimedia Conference (pp. 1029-1031). Association for Computing Machinery. https://doi.org/10.1145/3343031.3350580

@inproceedings{d6289e9c7fa44aa991ed67472d6a33bf,

title = "Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images",

abstract = "In this demonstration, we present Exquisitor, a media explorer capable of learning user preferences in real-time during interactions with the 99.2 million images of YFCC100M. Exquisitor owes its efficiency to innovations in data representation, compression, and indexing. Exquisitor can complete each interaction round, including learning preferences and presenting the most relevant results, in less than 30 ms using only a single CPU core and modest RAM. In short, Exquisitor can bring large-scale interactive learning to standard desktops and laptops, and even high-end mobile devices.",

keywords = "Interactive multimodal learning, Scalability, 100 million images, Interactive multimodal learning, Scalability, 100 million images",

author = "Hanna Ragnarsd{\'o}ttir and {\TH}{\'o}rhildur {\TH}orleiksd{\'o}ttir and Khan, {Omar Shahbaz} and J{\'o}nsson, {Bj{\"o}rn Th{\'o}r} and Gu{\dh}mundsson, {Gylfi {\TH}{\'o}r} and Jan Zah{\'a}lka and Stevan Rudinac and Laurent Amsaleg and Marcel Worring",

year = "2019",

month = oct,

doi = "10.1145/3343031.3350580",

language = "English",

pages = "1029--1031",

booktitle = "Proceedings of the ACM Multimedia Conference",

publisher = "Association for Computing Machinery",

address = "United States",

}

Ragnarsdóttir, H, Þorleiksdóttir, Þ, Khan, OS, Jónsson, BT, Guðmundsson, GÞ, Zahálka, J, Rudinac, S, Amsaleg, L & Worring, M 2019, Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images. in Proceedings of the ACM Multimedia Conference. Association for Computing Machinery, Nice, France, pp. 1029-1031. https://doi.org/10.1145/3343031.3350580

Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images. / Ragnarsdóttir, Hanna; Þorleiksdóttir, Þórhildur; Khan, Omar Shahbaz et al.
Proceedings of the ACM Multimedia Conference. Nice, France: Association for Computing Machinery, 2019. p. 1029-1031.

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images

AU - Ragnarsdóttir, Hanna

AU - Þorleiksdóttir, Þórhildur

AU - Khan, Omar Shahbaz

AU - Jónsson, Björn Thór

AU - Guðmundsson, Gylfi Þór

AU - Zahálka, Jan

AU - Rudinac, Stevan

AU - Amsaleg, Laurent

AU - Worring, Marcel

PY - 2019/10

Y1 - 2019/10

N2 - In this demonstration, we present Exquisitor, a media explorer capable of learning user preferences in real-time during interactions with the 99.2 million images of YFCC100M. Exquisitor owes its efficiency to innovations in data representation, compression, and indexing. Exquisitor can complete each interaction round, including learning preferences and presenting the most relevant results, in less than 30 ms using only a single CPU core and modest RAM. In short, Exquisitor can bring large-scale interactive learning to standard desktops and laptops, and even high-end mobile devices.

AB - In this demonstration, we present Exquisitor, a media explorer capable of learning user preferences in real-time during interactions with the 99.2 million images of YFCC100M. Exquisitor owes its efficiency to innovations in data representation, compression, and indexing. Exquisitor can complete each interaction round, including learning preferences and presenting the most relevant results, in less than 30 ms using only a single CPU core and modest RAM. In short, Exquisitor can bring large-scale interactive learning to standard desktops and laptops, and even high-end mobile devices.

KW - Interactive multimodal learning

KW - Scalability

KW - 100 million images

KW - Interactive multimodal learning

KW - Scalability

KW - 100 million images

U2 - 10.1145/3343031.3350580

DO - 10.1145/3343031.3350580

M3 - Article in proceedings

SP - 1029

EP - 1031

BT - Proceedings of the ACM Multimedia Conference

PB - Association for Computing Machinery

CY - Nice, France

ER -

Exquisitor: Breaking the Interaction Barrier for Exploration of 100 Million Images

Abstract

Keywords

Access to Document

Fingerprint

Cite this