Combining spatial proximity and temporal continuity for learning invariant representations

Olcay Kursun; Tevfik Aytekin

doi:10.1109/ASONAM.2012.157

Combining spatial proximity and temporal continuity for learning invariant representations

Olcay Kursun, Tevfik Aytekin

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

1 Citation (Scopus)

Abstract

Location and time are two critical aspects of most security-related events, and thus, spatiotemporal data analysis plays a central role in many security-related applications. The human brain has great capabilities of developing invariant representations of objects by taking advantage of both spatial similarity of features of objects/events and their relative timings (temporal information). Trace learning rule is one well-known solution for this problem of combining temporal relations with spatial proximity in clustering tasks such as the one performed by self organizing maps. In this work, we investigate a two stage mechanism: i) finding local clusters using spatial proximity, ii) grouping these clusters as suggested by temporal continuity patterns. We show our experimental results on a movie created from face images.

Original language	English
Title of host publication	Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012
Pages	871-873
Number of pages	3
DOIs	https://doi.org/10.1109/ASONAM.2012.157
Publication status	Published - 2012
Externally published	Yes
Event	2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012 - Istanbul, Turkey Duration: 26 Aug 2012 → 29 Aug 2012

Publication series

Name	Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

Conference

Conference	2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012
Country/Territory	Turkey
City	Istanbul
Period	26/08/12 → 29/08/12

Keywords

Face recognition
Invariant feature extraction
ORL face-movie
Self-organizing maps (SOM)
Spatiotemporal clustering
Trace learning rule

Access to Document

10.1109/ASONAM.2012.157

Cite this

Kursun, O., & Aytekin, T. (2012). Combining spatial proximity and temporal continuity for learning invariant representations. In Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012 (pp. 871-873). Article 6425650 (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012). https://doi.org/10.1109/ASONAM.2012.157

Kursun, Olcay ; Aytekin, Tevfik. / Combining spatial proximity and temporal continuity for learning invariant representations. Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. pp. 871-873 (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012).

@inproceedings{346b97f9390b4f9a9a66266f17a8e54b,

title = "Combining spatial proximity and temporal continuity for learning invariant representations",

abstract = "Location and time are two critical aspects of most security-related events, and thus, spatiotemporal data analysis plays a central role in many security-related applications. The human brain has great capabilities of developing invariant representations of objects by taking advantage of both spatial similarity of features of objects/events and their relative timings (temporal information). Trace learning rule is one well-known solution for this problem of combining temporal relations with spatial proximity in clustering tasks such as the one performed by self organizing maps. In this work, we investigate a two stage mechanism: i) finding local clusters using spatial proximity, ii) grouping these clusters as suggested by temporal continuity patterns. We show our experimental results on a movie created from face images.",

keywords = "Face recognition, Invariant feature extraction, ORL face-movie, Self-organizing maps (SOM), Spatiotemporal clustering, Trace learning rule",

author = "Olcay Kursun and Tevfik Aytekin",

year = "2012",

doi = "10.1109/ASONAM.2012.157",

language = "English",

isbn = "9780769547992",

series = "Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012",

pages = "871--873",

booktitle = "Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012",

note = "2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012 ; Conference date: 26-08-2012 Through 29-08-2012",

}

Kursun, O & Aytekin, T 2012, Combining spatial proximity and temporal continuity for learning invariant representations. in Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012., 6425650, Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012, pp. 871-873, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012, Istanbul, Turkey, 26/08/12. https://doi.org/10.1109/ASONAM.2012.157

Combining spatial proximity and temporal continuity for learning invariant representations. / Kursun, Olcay; Aytekin, Tevfik.
Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. p. 871-873 6425650 (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012).

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

TY - GEN

T1 - Combining spatial proximity and temporal continuity for learning invariant representations

AU - Kursun, Olcay

AU - Aytekin, Tevfik

PY - 2012

Y1 - 2012

N2 - Location and time are two critical aspects of most security-related events, and thus, spatiotemporal data analysis plays a central role in many security-related applications. The human brain has great capabilities of developing invariant representations of objects by taking advantage of both spatial similarity of features of objects/events and their relative timings (temporal information). Trace learning rule is one well-known solution for this problem of combining temporal relations with spatial proximity in clustering tasks such as the one performed by self organizing maps. In this work, we investigate a two stage mechanism: i) finding local clusters using spatial proximity, ii) grouping these clusters as suggested by temporal continuity patterns. We show our experimental results on a movie created from face images.

AB - Location and time are two critical aspects of most security-related events, and thus, spatiotemporal data analysis plays a central role in many security-related applications. The human brain has great capabilities of developing invariant representations of objects by taking advantage of both spatial similarity of features of objects/events and their relative timings (temporal information). Trace learning rule is one well-known solution for this problem of combining temporal relations with spatial proximity in clustering tasks such as the one performed by self organizing maps. In this work, we investigate a two stage mechanism: i) finding local clusters using spatial proximity, ii) grouping these clusters as suggested by temporal continuity patterns. We show our experimental results on a movie created from face images.

KW - Face recognition

KW - Invariant feature extraction

KW - ORL face-movie

KW - Self-organizing maps (SOM)

KW - Spatiotemporal clustering

KW - Trace learning rule

UR - http://www.scopus.com/inward/record.url?scp=84874225214&partnerID=8YFLogxK

U2 - 10.1109/ASONAM.2012.157

DO - 10.1109/ASONAM.2012.157

M3 - Conference contribution

AN - SCOPUS:84874225214

SN - 9780769547992

T3 - Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

SP - 871

EP - 873

BT - Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

T2 - 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012

Y2 - 26 August 2012 through 29 August 2012

ER -

Kursun O, Aytekin T. Combining spatial proximity and temporal continuity for learning invariant representations. In Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012. 2012. p. 871-873. 6425650. (Proceedings of the 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2012). doi: 10.1109/ASONAM.2012.157

Combining spatial proximity and temporal continuity for learning invariant representations

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this