BERT-VQA: Visual Question Answering on Plots

TL;DR

This paper introduces BERT-VQA, a model for visual question answering on plots, exploring the effectiveness of cross-modality modules in VisualBERT-based architectures for this specific task.

Contribution

The study develops a BERT-VQA model with a pretrained ResNet encoder and evaluates its performance against a baseline, providing insights into model architecture effectiveness for plot VQA.

Findings

Cross-modality module in VisualBERT is not essential for plot VQA.

The proposed model offers a new approach to visual question answering on plots.

Insights into the challenges and architecture choices for plot-based VQA.

Abstract

Visual question answering has been an exciting challenge in the field of natural language understanding, as it requires deep learning models to exchange information from both vision and language domains. In this project, we aim to tackle a subtask of this problem, namely visual question answering on plots. To achieve this, we developed BERT-VQA, a VisualBERT-based model architecture with a pretrained ResNet 101 image encoder, along with a potential addition of joint fusion. We trained and evaluated this model against a baseline that consisted of a LSTM, a CNN, and a shallow classifier. The final outcome disproved our core hypothesis that the cross-modality module in VisualBERT is essential in aligning plot components with question phrases. Therefore, our work provided valuable insights into the difficulty of the plot question answering challenge as well as the appropriateness of different model architectures in solving this problem.