MSG-Chart: Multimodal Scene Graph for ChartQA

TL;DR

This paper introduces MSG-Chart, a multimodal scene graph approach for ChartQA that explicitly models relationships between chart elements and patterns, improving understanding and answering accuracy.

Contribution

It proposes a joint multimodal scene graph combining visual and textual graphs, enhancing chart understanding in a way that can be integrated with vision transformers.

Findings

Improved performance on ChartQA benchmark.

Effective integration with vision transformers.

Enhanced structural and semantic understanding of charts.

Abstract

Automatic Chart Question Answering (ChartQA) is challenging due to the complex distribution of chart elements with patterns of the underlying data not explicitly displayed in charts. To address this challenge, we design a joint multimodal scene graph for charts to explicitly represent the relationships between chart elements and their patterns. Our proposed multimodal scene graph includes a visual graph and a textual graph to jointly capture the structural and semantical knowledge from the chart. This graph module can be easily integrated with different vision transformers as inductive bias. Our experiments demonstrate that incorporating the proposed graph module enhances the understanding of charts' elements' structure and semantics, thereby improving performance on publicly available benchmarks, ChartQA and OpenCQA.