SciPostLayout: A Dataset for Layout Analysis and Layout Generation of Scientific Posters

TL;DR

The paper introduces SciPostLayout, a large dataset of scientific posters and papers with annotations, enabling research in layout analysis and generation, and demonstrates the challenges and potential of using large language models for automatic poster creation.

Contribution

We created and publicly released SciPostLayout, a comprehensive dataset for scientific poster layout analysis and generation, and evaluated existing models and LLMs on this new resource.

Findings

Poster layout analysis and generation are more challenging than with scientific papers.

Existing computer vision models show limited performance on poster layout tasks.

LLMs have potential for automatic scientific poster generation.

Abstract

Scientific posters are used to present the contributions of scientific papers effectively in a graphical format. However, creating a well-designed poster that efficiently summarizes the core of a paper is both labor-intensive and time-consuming. A system that can automatically generate well-designed posters from scientific papers would reduce the workload of authors and help readers understand the outline of the paper visually. Despite the demand for poster generation systems, only a limited research has been conduced due to the lack of publicly available datasets. Thus, in this study, we built the SciPostLayout dataset, which consists of 7,855 scientific posters and manual layout annotations for layout analysis and generation. SciPostLayout also contains 100 scientific papers paired with the posters. All of the posters and papers in our dataset are under the CC-BY license and are publicly available. As benchmark tests for the collected dataset, we conducted experiments for layout analysis and generation utilizing existing computer vision models and found that both layout analysis and generation of posters using SciPostLayout are more challenging than with scientific papers. We also conducted experiments on generating layouts from scientific papers to demonstrate the potential of utilizing LLM as a scientific poster generation system. The dataset is publicly available at https://huggingface.co/datasets/omron-sinicx/scipostlayout_v2. The code is also publicly available at https://github.com/omron-sinicx/scipostlayout.