Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers

Omar Shahbaz Khan; Aaron Duane; Björn Thór Jónsson; Jan Zahálka; Stevan Rudinac; Marcel Worring

doi:10.1145/3463948.3469255

Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers

Omar Shahbaz Khan, Aaron Duane, Björn Thór Jónsson, Jan Zahálka, Stevan Rudinac, Marcel Worring

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

Abstract

Exquisitor is a scalable media exploration system based on interactive learning. To satisfy a user's information need, the system asks the user for feedback on media items and uses that feedback to interactively construct a classifier, that is in turn used to identify the next potentially relevant set of media items. To facilitate effective exploration of a collection, the system offers filters to narrow the scope of exploration, search functionality for finding good examples for the classifier, and support for timeline browsing of videos or image sequences. For this year's Lifelog Search Challenge, we have enhanced Exquisitor to better support tasks with a temporal component, by adding features that allow the user to build multiple classifiers and merge the classifier results, using both traditional set operators and advanced temporal operators.

Original language	English
Title of host publication	LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge
Number of pages	4
Place of Publication	Taipei, Taiwan (virtual)
Publisher	Association for Computing Machinery
Publication date	Aug 2021
Pages	3-6
DOIs	https://doi.org/10.1145/3463948.3469255 https://doi.org/10.1145/3463948.3469255
Publication status	Published - Aug 2021

Keywords

Scalable media exploration
Interactive learning
User feedback
Media classification
Temporal operators

Access to Document

Exquisitor__LSC_2021Accepted author manuscript, 13.7 MB

Cite this

Khan, O. S., Duane, A., Jónsson, B. T., Zahálka, J., Rudinac, S., & Worring, M. (2021). Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers. In LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge (pp. 3-6). Association for Computing Machinery. https://doi.org/10.1145/3463948.3469255, https://doi.org/10.1145/3463948.3469255

@inproceedings{c0a27faee7884cc8bcb038ae1c739a7f,

title = "Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers",

abstract = "Exquisitor is a scalable media exploration system based on interactive learning. To satisfy a user's information need, the system asks the user for feedback on media items and uses that feedback to interactively construct a classifier, that is in turn used to identify the next potentially relevant set of media items. To facilitate effective exploration of a collection, the system offers filters to narrow the scope of exploration, search functionality for finding good examples for the classifier, and support for timeline browsing of videos or image sequences. For this year's Lifelog Search Challenge, we have enhanced Exquisitor to better support tasks with a temporal component, by adding features that allow the user to build multiple classifiers and merge the classifier results, using both traditional set operators and advanced temporal operators.",

keywords = "Scalable media exploration, , Interactive learning, , User feedback, , Media classification, , Temporal operators, Scalable media exploration, , Interactive learning, , User feedback, , Media classification, , Temporal operators",

author = "Khan, {Omar Shahbaz} and Aaron Duane and J{\'o}nsson, {Bj{\"o}rn Th{\'o}r} and Jan Zah{\'a}lka and Stevan Rudinac and Marcel Worring",

year = "2021",

month = aug,

doi = "10.1145/3463948.3469255",

language = "English",

pages = "3--6",

booktitle = "LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge",

publisher = "Association for Computing Machinery",

address = "United States",

}

Khan, OS, Duane, A, Jónsson, BT, Zahálka, J, Rudinac, S & Worring, M 2021, Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers. in LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge. Association for Computing Machinery, Taipei, Taiwan (virtual), pp. 3-6. https://doi.org/10.1145/3463948.3469255, https://doi.org/10.1145/3463948.3469255

Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers. / Khan, Omar Shahbaz; Duane, Aaron; Jónsson, Björn Thór et al.
LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge. Taipei, Taiwan (virtual): Association for Computing Machinery, 2021. p. 3-6.

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

TY - GEN

T1 - Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers

AU - Khan, Omar Shahbaz

AU - Duane, Aaron

AU - Jónsson, Björn Thór

AU - Zahálka, Jan

AU - Rudinac, Stevan

AU - Worring, Marcel

PY - 2021/8

Y1 - 2021/8

N2 - Exquisitor is a scalable media exploration system based on interactive learning. To satisfy a user's information need, the system asks the user for feedback on media items and uses that feedback to interactively construct a classifier, that is in turn used to identify the next potentially relevant set of media items. To facilitate effective exploration of a collection, the system offers filters to narrow the scope of exploration, search functionality for finding good examples for the classifier, and support for timeline browsing of videos or image sequences. For this year's Lifelog Search Challenge, we have enhanced Exquisitor to better support tasks with a temporal component, by adding features that allow the user to build multiple classifiers and merge the classifier results, using both traditional set operators and advanced temporal operators.

AB - Exquisitor is a scalable media exploration system based on interactive learning. To satisfy a user's information need, the system asks the user for feedback on media items and uses that feedback to interactively construct a classifier, that is in turn used to identify the next potentially relevant set of media items. To facilitate effective exploration of a collection, the system offers filters to narrow the scope of exploration, search functionality for finding good examples for the classifier, and support for timeline browsing of videos or image sequences. For this year's Lifelog Search Challenge, we have enhanced Exquisitor to better support tasks with a temporal component, by adding features that allow the user to build multiple classifiers and merge the classifier results, using both traditional set operators and advanced temporal operators.

KW - Scalable media exploration

KW - Interactive learning

KW - User feedback

KW - Media classification

KW - Temporal operators

KW - Scalable media exploration

KW - Interactive learning

KW - User feedback

KW - Media classification

KW - Temporal operators

U2 - 10.1145/3463948.3469255

DO - 10.1145/3463948.3469255

M3 - Article in proceedings

SP - 3

EP - 6

BT - LSC '21: Proceedings of the 4th Annual on Lifelog Search Challenge

PB - Association for Computing Machinery

CY - Taipei, Taiwan (virtual)

ER -

Exquisitor at the Lifelog Search Challenge 2021: Relationships Between Semantic Classifiers

Abstract

Keywords

Access to Document

Fingerprint

Cite this