Intelligent Grimm -- Open-ended Visual Storytelling via Latent Diffusion Models

TL;DR

This paper introduces StoryGen, a novel auto-regressive model for open-ended visual storytelling that generates coherent image sequences from storylines, supported by a large-scale dataset and validated through experiments and human evaluations.

Contribution

The paper presents a new model, a large dataset, and a pipeline for visual storytelling, advancing the coherence and diversity of generated image sequences.

Findings

StoryGen outperforms baselines in coherence and character consistency

The dataset StorySalon enables diverse storytelling scenarios

Model generalizes to unseen characters without additional optimization

Abstract

Generative models have recently exhibited exceptional capabilities in text-to-image generation, but still struggle to generate image sequences coherently. In this work, we focus on a novel, yet challenging task of generating a coherent image sequence based on a given storyline, denoted as open-ended visual storytelling. We make the following three contributions: (i) to fulfill the task of visual storytelling, we propose a learning-based auto-regressive image generation model, termed as StoryGen, with a novel vision-language context module, that enables to generate the current frame by conditioning on the corresponding text prompt and preceding image-caption pairs; (ii) to address the data shortage of visual storytelling, we collect paired image-text sequences by sourcing from online videos and open-source E-books, establishing processing pipeline for constructing a large-scale dataset with diverse characters, storylines, and artistic styles, named StorySalon; (iii) Quantitative experiments and human evaluations have validated the superiority of our StoryGen, where we show StoryGen can generalize to unseen characters without any optimization, and generate image sequences with coherent content and consistent character. Code, dataset, and models are available at https://haoningwu3639.github.io/StoryGen_Webpage/