Corpora
Visual Question Answering
Open-ended questions about images. These questions require an understanding of vision, language and commonsense knowledge to answer.
Introduced in:
Antol, S. and Agrawal, A. and Lu, J. and Mitchell, M. and Batra, D. and Zitnick, C. L. and Parikh, D. (2015). VQA: Visual Question Answering.
International Conference on Computer Vision (ICCV).
The Visual Storying Storytelling Dataset
Conceptual, Abstract Language in Temporal Sequence.
Used in:
Huang, T.-H. and Ferraro, F. and Mostafazadeh, N. and Misra, I. and Agrawal, A. and Devlin, J. and Girshick, R. and He, X. and Kohli, P. and Batra, D. and Zitnick, C. L. and Parikh, D. and Vanderwende, L. and Galley, M. and Mitchell, M. (2016). Visual Storytelling. Proceedings of NAACL 2016.
Collected dialogues:
Collected references to visible objects:
GRE3D3 Corpus, XML format
http://www.m-mitchell.com/corpora/GRE3D3_corpus.
Collected by Jette Viethen; this version has just converted it to XML format.
Discussed in:
Viethen, J., Mitchell, M., and Krahmer, E. Graphs and Spatial Relations in the Generation of Referring Expressions. Proceedings of ENLG 2013.
The Typicality Corpus
http://www.m-mitchell.com/corpora/typicality_corpus.
Discussed in:
Mitchell, M., Reiter, E., and van Deemter, K. (2013). Typicality and Object Reference. Proceedings of CogSci 2013.
The Typicality Corpus by Margaret Mitchell is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Based on a work at The University of Aberdeen.
The Craft Corpus
http://www.m-mitchell.com/corpora/craft_corpus.
Discussed in:
Mitchell, M., van Deemter, K., and Reiter, E. (2010). Natural Reference to Objects in a Visual Domain. Proceedings of INLG 2010.
The Craft Corpus by Margaret Mitchell is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Based on a work at The University of Aberdeen.
The Size Corpus
http://www.m-mitchell.com/corpora/size_corpus.
Discussed in:
Mitchell, M., van Deemter, K., and Reiter, E. (2011). Applying Machine Learning to the Choice of Size Modifiers. Proceedings of PRE-CogSci 2011.
The Size Corpus by Margaret Mitchell is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
Based on a work at The University of Aberdeen.
Please contact me if you wish to use this data beyond the scope of the Creative Commons license. (E-mail address at top of page).