Focal Visual-Text Attention for Visual Question Answering

TL;DR

The paper introduces FVTA, a neural network that enhances visual question answering over multimedia collections by dynamically focusing on relevant media snippets and providing justifications, achieving state-of-the-art results.

Contribution

It proposes a novel end-to-end hierarchical neural network for collective reasoning in visual question answering involving sequences of images and text.

Findings

Achieves state-of-the-art on MemexQA dataset.

Provides justifications for answers.

Performs competitively on MovieQA dataset.

Abstract

Recent insights on language and vision with neural networks have been successfully applied to simple single-image visual question answering. However, to tackle real-life question answering problems on multimedia collections such as personal photos, we have to look at whole collections with sequences of photos or videos. When answering questions from a large collection, a natural problem is to identify snippets to support the answer. In this paper, we describe a novel neural network called Focal Visual-Text Attention network (FVTA) for collective reasoning in visual question answering, where both visual and text sequence information such as images and text metadata are presented. FVTA introduces an end-to-end approach that makes use of a hierarchical process to dynamically determine what media and what time to focus on in the sequential data to answer the question. FVTA can not only answer the questions well but also provides the justifications which the system results are based upon to get the answers. FVTA achieves state-of-the-art performance on the MemexQA dataset and competitive results on the MovieQA dataset.