Constructing spatial representations of variable detail for sketch recognition

Andrew Lovett; Morteza Dehghani; Kenneth D Forbus

Constructing spatial representations of variable detail for sketch recognition

Andrew Lovett^*, Morteza Dehghani, Kenneth D Forbus

^*Corresponding author for this work

Computer Science

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

2 Scopus citations

Abstract

We describe a system which constructs spatial representations of sketches drawn by users. These representations are currently being used as the input for a spatial reasoning system which learns classifiers for performing sketch recognition. The spatial reasoning system requires representations at a level of detail sparser than that which the representation constructor normally builds. Therefore, we describe how the representation constructor ranks the expressions in its output so that the number of expressions in the representation can be decreased with minimal loss of information. We evaluate the overall system, showing that it is able to learn and utilize classifiers for complex sketches even when the representation size is sharply diminished.

Original language	English (US)
Title of host publication	Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems - Papers from the 2007 AAAI Spring Symposium, Technical Report
Pages	35-40
Number of pages	6
Volume	SS-07-01
State	Published - Dec 28 2007
Event	2007 AAAI Spring Symposium - Stanford, CA, United States Duration: Mar 26 2007 → Mar 28 2007

Other

Other	2007 AAAI Spring Symposium
Country	United States
City	Stanford, CA
Period	3/26/07 → 3/28/07

ASJC Scopus subject areas

Engineering(all)

Fingerprint Dive into the research topics of 'Constructing spatial representations of variable detail for sketch recognition'. Together they form a unique fingerprint.

Cite this

@inproceedings{e192441b7d0b40029db4614af3f66767,

title = "Constructing spatial representations of variable detail for sketch recognition",

abstract = "We describe a system which constructs spatial representations of sketches drawn by users. These representations are currently being used as the input for a spatial reasoning system which learns classifiers for performing sketch recognition. The spatial reasoning system requires representations at a level of detail sparser than that which the representation constructor normally builds. Therefore, we describe how the representation constructor ranks the expressions in its output so that the number of expressions in the representation can be decreased with minimal loss of information. We evaluate the overall system, showing that it is able to learn and utilize classifiers for complex sketches even when the representation size is sharply diminished.",

author = "Andrew Lovett and Morteza Dehghani and Forbus, {Kenneth D}",

year = "2007",

month = dec,

day = "28",

language = "English (US)",

isbn = "9781577353102",

volume = "SS-07-01",

pages = "35--40",

booktitle = "Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems - Papers from the 2007 AAAI Spring Symposium, Technical Report",

note = "2007 AAAI Spring Symposium ; Conference date: 26-03-2007 Through 28-03-2007",

}

Constructing spatial representations of variable detail for sketch recognition. / Lovett, Andrew; Dehghani, Morteza; Forbus, Kenneth D.

Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems - Papers from the 2007 AAAI Spring Symposium, Technical Report. Vol. SS-07-01 2007. p. 35-40.

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

TY - GEN

T1 - Constructing spatial representations of variable detail for sketch recognition

AU - Lovett, Andrew

AU - Dehghani, Morteza

AU - Forbus, Kenneth D

PY - 2007/12/28

Y1 - 2007/12/28

N2 - We describe a system which constructs spatial representations of sketches drawn by users. These representations are currently being used as the input for a spatial reasoning system which learns classifiers for performing sketch recognition. The spatial reasoning system requires representations at a level of detail sparser than that which the representation constructor normally builds. Therefore, we describe how the representation constructor ranks the expressions in its output so that the number of expressions in the representation can be decreased with minimal loss of information. We evaluate the overall system, showing that it is able to learn and utilize classifiers for complex sketches even when the representation size is sharply diminished.

AB - We describe a system which constructs spatial representations of sketches drawn by users. These representations are currently being used as the input for a spatial reasoning system which learns classifiers for performing sketch recognition. The spatial reasoning system requires representations at a level of detail sparser than that which the representation constructor normally builds. Therefore, we describe how the representation constructor ranks the expressions in its output so that the number of expressions in the representation can be decreased with minimal loss of information. We evaluate the overall system, showing that it is able to learn and utilize classifiers for complex sketches even when the representation size is sharply diminished.

UR - http://www.scopus.com/inward/record.url?scp=37349080928&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37349080928&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:37349080928

SN - 9781577353102

VL - SS-07-01

SP - 35

EP - 40

BT - Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems - Papers from the 2007 AAAI Spring Symposium, Technical Report

T2 - 2007 AAAI Spring Symposium

Y2 - 26 March 2007 through 28 March 2007

ER -