Multimodal Differential Network for Visual Question Generation [Paper: EMNLP-2018]

Badri N. Patro, Sandeep Kumar, Vinod K. Kurmi, Vinay P. Namboodiri




Statistical Significance Analysis

The mean rank of all the models on the basis of METEOR and Blue score are plotted on the x-axis. Here Joint refers to our MDN-Joint model. The colored lines between the two models represents that these models are not significantly different from each other. METEOR Score Blue Score

Human Survey

For human evalutions, we have created 10 different google forms each have 10 random set of image-question pairs. The link for the human evalation form is given as: Human Survey

Perceptual Realism

Perceptual Realism Plot for human survey. Here every question has different number of responses and hence the threshold which is the half of total responses for each question is varying. This plot is only for 50 of the 100 questions involved in the survey.

Perceptual Realism



  author = 	"Patro, Badri Narayana
		and Kumar, Sandeep
		and Kurmi, Vinod Kumar
		and Namboodiri, Vinay",
  title = 	"Multimodal Differential Network for Visual Question Generation",
  booktitle = 	"Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
  year = 	"2018",
  publisher = 	"Association for Computational Linguistics",
  pages = 	"4002--4012",
  location = 	"Brussels, Belgium",
  url = 	""