@inproceedings{ebrahimikahou2017figureqa,
author = {Ebrahimi Kahou, Samira and Atkinson, Adam and Michalski, Vincent and Kadar, Akos and Trischler, Adam and Bengio, Yoshua},
title = {FigureQA: An annotated figure dataset for visual reasoning},
booktitle = {Visually grounded interaction and language workshop, NIPS 2017},
year = {2017},
month = {December},
abstract = {We introduce FigureQA, a visual reasoning corpus of over one million question-answer pairs grounded in over 100,000 images. The images are synthetic, scientific-style figures from five classes: line plots, dot-line plots, vertical and horizontal bar graphs, and pie charts. We formulate our reasoning task by generating questions from 15 templates; questions concern various relationships between plot elements and examine characteristics like the maximum, the minimum, area-under-the-curve, smoothness, and intersection. To resolve, such questions often require reference to multiple plot elements and synthesis of information distributed spatially throughout a figure. To facilitate the training of machine learning systems, the corpus also includes side data that can be used to formulate auxiliary objectives. In particular, we provide the numerical data used to generate each figure as well as bounding-box annotations for all plot elements. We study the proposed visual reasoning task by training several models, including the recently proposed Relation Network as strong baseline. Preliminary results indicate that the task poses a significant machine learning challenge. We envision FigureQA as a first step towards developing models that can intuitively recognize patterns from visual representations of data.},
url = {http://approjects.co.za/?big=en-us/research/publication/figureqa-an-annotated-figure-dataset-for-visual-reasoning/},
}