Advanced

Linking Entities Across Images and Text

Weegar, Rebecka; Åström, Karl LU and Nugues, Pierre LU (2015) Nineteenth Conference on Computational Natural Language Learning (CoNLL 2015)
Abstract
This paper describes a set of methods to link entities across images and text. As a corpus, we used a data set of images,

where each image is commented by a short caption and where the regions in the images are manually segmented and labeled with a category. We extracted the entity mentions from the captions and we computed a semantic similarity between the mentions and the region labels. We also

measured the statistical associations between these mentions and the labels and we combined them with the semantic similarity to produce mappings in the form of pairs consisting of a region label and

a caption entity. In a second step, we used the syntactic relationships between the mentions and the spatial... (More)
This paper describes a set of methods to link entities across images and text. As a corpus, we used a data set of images,

where each image is commented by a short caption and where the regions in the images are manually segmented and labeled with a category. We extracted the entity mentions from the captions and we computed a semantic similarity between the mentions and the region labels. We also

measured the statistical associations between these mentions and the labels and we combined them with the semantic similarity to produce mappings in the form of pairs consisting of a region label and

a caption entity. In a second step, we used the syntactic relationships between the mentions and the spatial relationships

between the regions to rerank the lists of candidate mappings. To evaluate our methods, we annotated a test set of 200 images, where we manually linked the im- age regions to their corresponding mentions in the captions. Eventually, we could match objects in pictures to their correct mentions for nearly 89 percent of the segments, when such a matching exists. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
Nineteenth Conference on Computational Natural Language Learning (CoNLL 2015)
language
English
LU publication?
yes
id
eafeeff2-deb2-4739-96a1-b0b6927929f4 (old id 7864937)
alternative location
http://www.aclweb.org/anthology/K/K15/K15-1019.pdf
date added to LUP
2015-09-14 10:17:46
date last changed
2016-04-16 12:09:39
@misc{eafeeff2-deb2-4739-96a1-b0b6927929f4,
  abstract     = {This paper describes a set of methods to link entities across images and text. As a corpus, we used a data set of images,<br/><br>
where each image is commented by a short caption and where the regions in the images are manually segmented and labeled with a category. We extracted the entity mentions from the captions and we computed a semantic similarity between the mentions and the region labels. We also<br/><br>
measured the statistical associations between these mentions and the labels and we combined them with the semantic similarity to produce mappings in the form of pairs consisting of a region label and<br/><br>
a caption entity. In a second step, we used the syntactic relationships between the mentions and the spatial relationships<br/><br>
between the regions to rerank the lists of candidate mappings. To evaluate our methods, we annotated a test set of 200 images, where we manually linked the im- age regions to their corresponding mentions in the captions. Eventually, we could match objects in pictures to their correct mentions for nearly 89 percent of the segments, when such a matching exists.},
  author       = {Weegar, Rebecka and Åström, Karl and Nugues, Pierre},
  language     = {eng},
  title        = {Linking Entities Across Images and Text},
  year         = {2015},
}