Disambiguating Visual Verbs
- Submitting institution
-
University of Edinburgh
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 58512519
- Type
- D - Journal article
- DOI
-
10.1109/TPAMI.2017.2786699
- Title of journal
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Article number
- -
- First page
- 311
- Volume
- 41
- Issue
- 2
- ISSN
- 0162-8828
- Open access status
- Compliant
- Month of publication
- December
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
2
- Research group(s)
-
D - Language, Interaction and Robotics
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Word sense disambiguation is a standard task in natural language processing. This paper proposes a novel, more challenging task, visual sense disambiguation: given an image and a verb, assign the correct verb sense. Both supervised and unsupervised models for the task are introduced, based on textual, visual and multimodal embeddings. We also introduce a new dataset, which has subsequently been used by other researchers, e.g., forming the basis of a shared task at the Conference on Machine Translation (WMT 2018). Appeared in TPAMI, the premier journal in computer vision and machine learning.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -