Knowledge propagation in large image databases using neighborhood information

Michael E. Houle; Vincent Oria; Shin'ichi Satoh; Jichao Sun

doi:10.1145/2072298.2071931

Knowledge propagation in large image databases using neighborhood information

Michael E. Houle, Vincent Oria, Shin'ichi Satoh, Jichao Sun

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

The aim of this paper is to reduce to a minimum the level of human intervention in the semantic annotation process of images. Ideally, only one copy of each object of interest would be labeled manually, and the labels would then be propagated automatically to all other occurrences of the objects in the database. To that end, we propose a neighbor-based influence propagation approach KProp which builds a voting model and propagates the knowledge associated to some objects to similar objects. We show that KProp can perform efficiently through matrix computations and achieve better performance with fewer labeled examples per object.

Original language	English (US)
Title of host publication	MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops
Pages	1033-1036
Number of pages	4
DOIs	https://doi.org/10.1145/2072298.2071931
State	Published - Dec 29 2011
Externally published	Yes
Event	19th ACM International Conference on Multimedia ACM Multimedia 2011, MM'11 - Scottsdale, AZ, United States Duration: Nov 28 2011 → Dec 1 2011

Other

Other	19th ACM International Conference on Multimedia ACM Multimedia 2011, MM'11
Country	United States
City	Scottsdale, AZ
Period	11/28/11 → 12/1/11

All Science Journal Classification (ASJC) codes

Computer Graphics and Computer-Aided Design
Human-Computer Interaction

Access to Document

10.1145/2072298.2071931

Fingerprint Dive into the research topics of 'Knowledge propagation in large image databases using neighborhood information'. Together they form a unique fingerprint.

Cite this

@inproceedings{f2bb75d5bed54f039bf9004646bd9c2b,

title = "Knowledge propagation in large image databases using neighborhood information",

abstract = "The aim of this paper is to reduce to a minimum the level of human intervention in the semantic annotation process of images. Ideally, only one copy of each object of interest would be labeled manually, and the labels would then be propagated automatically to all other occurrences of the objects in the database. To that end, we propose a neighbor-based influence propagation approach KProp which builds a voting model and propagates the knowledge associated to some objects to similar objects. We show that KProp can perform efficiently through matrix computations and achieve better performance with fewer labeled examples per object.",

author = "Houle, {Michael E.} and Vincent Oria and Shin'ichi Satoh and Jichao Sun",

year = "2011",

month = dec,

day = "29",

doi = "10.1145/2072298.2071931",

language = "English (US)",

isbn = "9781450306164",

pages = "1033--1036",

booktitle = "MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops",

note = "19th ACM International Conference on Multimedia ACM Multimedia 2011, MM'11 ; Conference date: 28-11-2011 Through 01-12-2011",

}

Houle, ME, Oria, V, Satoh, S & Sun, J 2011, Knowledge propagation in large image databases using neighborhood information. in MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops. pp. 1033-1036, 19th ACM International Conference on Multimedia ACM Multimedia 2011, MM'11, Scottsdale, AZ, United States, 11/28/11. https://doi.org/10.1145/2072298.2071931

TY - GEN

T1 - Knowledge propagation in large image databases using neighborhood information

AU - Houle, Michael E.

AU - Oria, Vincent

AU - Satoh, Shin'ichi

AU - Sun, Jichao

PY - 2011/12/29

Y1 - 2011/12/29

N2 - The aim of this paper is to reduce to a minimum the level of human intervention in the semantic annotation process of images. Ideally, only one copy of each object of interest would be labeled manually, and the labels would then be propagated automatically to all other occurrences of the objects in the database. To that end, we propose a neighbor-based influence propagation approach KProp which builds a voting model and propagates the knowledge associated to some objects to similar objects. We show that KProp can perform efficiently through matrix computations and achieve better performance with fewer labeled examples per object.

AB - The aim of this paper is to reduce to a minimum the level of human intervention in the semantic annotation process of images. Ideally, only one copy of each object of interest would be labeled manually, and the labels would then be propagated automatically to all other occurrences of the objects in the database. To that end, we propose a neighbor-based influence propagation approach KProp which builds a voting model and propagates the knowledge associated to some objects to similar objects. We show that KProp can perform efficiently through matrix computations and achieve better performance with fewer labeled examples per object.

UR - http://www.scopus.com/inward/record.url?scp=84455202819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84455202819&partnerID=8YFLogxK

U2 - 10.1145/2072298.2071931

DO - 10.1145/2072298.2071931

M3 - Conference contribution

AN - SCOPUS:84455202819

SN - 9781450306164

SP - 1033

EP - 1036

BT - MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops

T2 - 19th ACM International Conference on Multimedia ACM Multimedia 2011, MM'11

Y2 - 28 November 2011 through 1 December 2011

ER -

Knowledge propagation in large image databases using neighborhood information

Abstract

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Cite this