Login (DCU Staff Only)
Login (DCU Staff Only)

DORAS | DCU Research Repository

Explore open access research and scholarly works from DCU

Advanced Search

Adding the third dimension to spatial relation detection in 2d images

Birmingham, Brandon orcid logoORCID: 0000-0002-3006-3526, Muscat, Adrian orcid logoORCID: 0000-0002-9157-2818 and Belz, Anya orcid logoORCID: 0000-0002-0552-8096 (2018) Adding the third dimension to spatial relation detection in 2d images. In: 11th International Conference on Natural Language Generation, 5-8 Nov 2018, Tilburg, The Netherlands.

Abstract
Detection of spatial relations between objects in images is currently a popular subject in image description research. A range of different language and geometric object features have been used in this context, but methods have not so far used explicit information about the third dimension (depth), except when manually added to annotations. The lack of such information hampers detection of spatial relations that are inherently 3D. In this paper, we use a fully automatic method for creating a depth map of an image and derive several different object-level depth features from it which we add to an existing feature set to test the effect on spatial relation detection. We show that performance increases are obtained from adding depth features in all scenarios tested.
Metadata
Item Type:Conference or Workshop Item (Paper)
Event Type:Conference
Refereed:Yes
Subjects:Computer Science > Computational linguistics
DCU Faculties and Centres:Research Initiatives and Centres > ADAPT
Published in: Krahmer, Emiel, Gatt, Albert and Goudbeek, Martijn, (eds.) Proceedings of the 11th International Conference on Natural Language Generation. . Association for Computational Linguistics (ACL).
Publisher:Association for Computational Linguistics (ACL)
Official URL:https://doi.org/10.18653/v1/W18-6517
Copyright Information:© 2018 Association for Computational Linguistics
ID Code:28624
Deposited On:07 Jul 2023 11:08 by Anya Belz . Last Modified 12 Jul 2023 11:13
Documents

Full text available as:

[thumbnail of W18-6517.pdf]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Creative Commons: Attribution 4.0
840kB
Downloads

Downloads

Downloads per month over past year

Archive Staff Only: edit this record