2026-10-11T00:37:42Z https://www.scidb.cn/oai

10.57760/sciencedb.39763 2026-06-12T14:12:25Z

2026-06-12 FableBench: A Structured Fable Dataset for Literary Story Visualization doi:10.57760/sciencedb.39763 en Story visualization is the process of transforming a continuous narrative into a series of sequentially arranged images. In current datasets such as PororoSV and FlintstonesSV, most of them are fixed animation segments with relatively few characters, short sentences, and consistent visual styles. Therefore, relying solely on these datasets to evaluate story visualization models may underestimate the difficulty brought by open literary texts. We have built a new dataset FableBench to address the above issues. This dataset selected 426 publicly available fables, and obtained 2788 scene level micro scenes through text collection, cleaning, narrative decomposition, prompt word organization, character visual archive extraction, and quality inspection. Each story contains a cleaned main text, symbolic tags, ordered micro scenes, Chinese scene descriptions, English image prompts, and structured character information. FableBench is mainly used for evaluation tasks related to the visualization of literary stories, and can support cross screen character consistency, image text alignment, prompt word engineering, and narrative decomposition. Preliminary experimental results indicate that fable texts can reveal some character drift and semantic compression issues that are not easily present in traditional animation datasets.  story visualization; scientific data; fable narrative; character consistency; text-to-image generation; evaluation dataset Shu Wenjun Wan Fucheng Xiao Yu Wang Rui PUBLIC https://creativecommons.org/licenses/by-nc-sa/4.0/ dataset Science Data Bank