RAID

A relation-augmented image descriptor

Paul Guerrero, Niloy J. Mitra, Peter Wonka

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

As humans, we regularly interpret scenes based on how objects are related, rather than based on the objects themselves. For example, we see a person riding an object X or a plank bridging two objects. Current methods provide limited support to search for content based on such relations. We present RAID, a relation-augmented image descriptor that supports queries based on inter-region relations. The key idea of our descriptor is to encode region-to-region relations as the spatial distribution of point-to-region relationships between two image regions. RAID allows sketch-based retrieval and requires minimal training data, thus making it suited even for querying uncommon relations. We evaluate the proposed descriptor by querying into large image databases and successfully extract nontrivial images demonstrating complex inter-region relations, which are easily missed or erroneously classified by existing methods. We assess the robustness of RAID on multiple datasets even when the region segmentation is computed automatically or very noisy.

Original languageEnglish (US)
Article numbera46
JournalACM Transactions on Graphics
Volume35
Issue number4
DOIs
StatePublished - Jul 11 2016

Fingerprint

Spatial distribution

Keywords

  • Image descriptors
  • Image retrieval
  • Relation-based query
  • Sketch-based query
  • Spatial relationships

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design

Cite this

RAID : A relation-augmented image descriptor. / Guerrero, Paul; Mitra, Niloy J.; Wonka, Peter.

In: ACM Transactions on Graphics, Vol. 35, No. 4, a46, 11.07.2016.

Research output: Contribution to journalArticle

Guerrero, Paul ; Mitra, Niloy J. ; Wonka, Peter. / RAID : A relation-augmented image descriptor. In: ACM Transactions on Graphics. 2016 ; Vol. 35, No. 4.
@article{ff44db3d72b54bab9a899ac0ac3e1fae,
title = "RAID: A relation-augmented image descriptor",
abstract = "As humans, we regularly interpret scenes based on how objects are related, rather than based on the objects themselves. For example, we see a person riding an object X or a plank bridging two objects. Current methods provide limited support to search for content based on such relations. We present RAID, a relation-augmented image descriptor that supports queries based on inter-region relations. The key idea of our descriptor is to encode region-to-region relations as the spatial distribution of point-to-region relationships between two image regions. RAID allows sketch-based retrieval and requires minimal training data, thus making it suited even for querying uncommon relations. We evaluate the proposed descriptor by querying into large image databases and successfully extract nontrivial images demonstrating complex inter-region relations, which are easily missed or erroneously classified by existing methods. We assess the robustness of RAID on multiple datasets even when the region segmentation is computed automatically or very noisy.",
keywords = "Image descriptors, Image retrieval, Relation-based query, Sketch-based query, Spatial relationships",
author = "Paul Guerrero and Mitra, {Niloy J.} and Peter Wonka",
year = "2016",
month = "7",
day = "11",
doi = "10.1145/2897824.2925939",
language = "English (US)",
volume = "35",
journal = "ACM Transactions on Graphics",
issn = "0730-0301",
publisher = "Association for Computing Machinery (ACM)",
number = "4",

}

TY - JOUR

T1 - RAID

T2 - A relation-augmented image descriptor

AU - Guerrero, Paul

AU - Mitra, Niloy J.

AU - Wonka, Peter

PY - 2016/7/11

Y1 - 2016/7/11

N2 - As humans, we regularly interpret scenes based on how objects are related, rather than based on the objects themselves. For example, we see a person riding an object X or a plank bridging two objects. Current methods provide limited support to search for content based on such relations. We present RAID, a relation-augmented image descriptor that supports queries based on inter-region relations. The key idea of our descriptor is to encode region-to-region relations as the spatial distribution of point-to-region relationships between two image regions. RAID allows sketch-based retrieval and requires minimal training data, thus making it suited even for querying uncommon relations. We evaluate the proposed descriptor by querying into large image databases and successfully extract nontrivial images demonstrating complex inter-region relations, which are easily missed or erroneously classified by existing methods. We assess the robustness of RAID on multiple datasets even when the region segmentation is computed automatically or very noisy.

AB - As humans, we regularly interpret scenes based on how objects are related, rather than based on the objects themselves. For example, we see a person riding an object X or a plank bridging two objects. Current methods provide limited support to search for content based on such relations. We present RAID, a relation-augmented image descriptor that supports queries based on inter-region relations. The key idea of our descriptor is to encode region-to-region relations as the spatial distribution of point-to-region relationships between two image regions. RAID allows sketch-based retrieval and requires minimal training data, thus making it suited even for querying uncommon relations. We evaluate the proposed descriptor by querying into large image databases and successfully extract nontrivial images demonstrating complex inter-region relations, which are easily missed or erroneously classified by existing methods. We assess the robustness of RAID on multiple datasets even when the region segmentation is computed automatically or very noisy.

KW - Image descriptors

KW - Image retrieval

KW - Relation-based query

KW - Sketch-based query

KW - Spatial relationships

UR - http://www.scopus.com/inward/record.url?scp=84980044688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84980044688&partnerID=8YFLogxK

U2 - 10.1145/2897824.2925939

DO - 10.1145/2897824.2925939

M3 - Article

VL - 35

JO - ACM Transactions on Graphics

JF - ACM Transactions on Graphics

SN - 0730-0301

IS - 4

M1 - a46

ER -