Visual Question Answering on Image Sets

TL;DR

This paper introduces the novel task of Image-Set Visual Question Answering (ISVQA), extending VQA to multiple images, and provides datasets, analysis, and baseline models to foster research in this area.

Contribution

It defines the ISVQA task, creates two new datasets for indoor and outdoor scenes, and analyzes dataset properties and baseline models for this multi-image VQA setting.

Findings

Datasets contain over 140,000 questions for image sets.

Analysis reveals biases and question-image dependencies.

Baseline models establish initial benchmarks for ISVQA.

Abstract

We introduce the task of Image-Set Visual Question Answering (ISVQA), which generalizes the commonly studied single-image VQA problem to multi-image settings. Taking a natural language question and a set of images as input, it aims to answer the question based on the content of the images. The questions can be about objects and relationships in one or more images or about the entire scene depicted by the image set. To enable research in this new topic, we introduce two ISVQA datasets - indoor and outdoor scenes. They simulate the real-world scenarios of indoor image collections and multiple car-mounted cameras, respectively. The indoor-scene dataset contains 91,479 human annotated questions for 48,138 image sets, and the outdoor-scene dataset has 49,617 questions for 12,746 image sets. We analyze the properties of the two datasets, including question-and-answer distributions, types of questions, biases in dataset, and question-image dependencies. We also build new baseline models to investigate new research challenges in ISVQA.