Exploring Models and Data for Image Question Answering
This addresses image QA for AI applications, with incremental improvements in models and data.
The paper tackled image-based question-answering by proposing neural networks and visual semantic embeddings, achieving a 1.8 times performance improvement over existing results on a dataset, and introduced a larger dataset via a question generation algorithm.
This work aims to address the problem of image-based question-answering (QA) with new models and datasets. In our work, we propose to use neural networks and visual semantic embeddings, without intermediate stages such as object detection and image segmentation, to predict answers to simple questions about images. Our model performs 1.8 times better than the only published results on an existing image QA dataset. We also present a question generation algorithm that converts image descriptions, which are widely available, into QA form. We used this algorithm to produce an order-of-magnitude larger dataset, with more evenly distributed answers. A suite of baseline results on this new dataset are also presented.