AutoPoster: A Highly Automatic and Content-aware Design System for Advertising Poster Generation

TL;DR

AutoPoster is an innovative system that automatically creates advertising posters by integrating content-aware design stages, leveraging a large annotated dataset, and producing visually harmonious posters with minimal input.

Contribution

The paper introduces AutoPoster, a fully automatic, content-aware poster generation system that utilizes a new large-scale dataset with visual attribute annotations.

Findings

AutoPoster produces aesthetically superior posters in user studies.

The system effectively integrates layout, tagline, and style prediction.

Experimental results demonstrate high quality and diversity of generated posters.

Abstract

Advertising posters, a form of information presentation, combine visual and linguistic modalities. Creating a poster involves multiple steps and necessitates design experience and creativity. This paper introduces AutoPoster, a highly automatic and content-aware system for generating advertising posters. With only product images and titles as inputs, AutoPoster can automatically produce posters of varying sizes through four key stages: image cleaning and retargeting, layout generation, tagline generation, and style attribute prediction. To ensure visual harmony of posters, two content-aware models are incorporated for layout and tagline generation. Moreover, we propose a novel multi-task Style Attribute Predictor (SAP) to jointly predict visual style attributes. Meanwhile, to our knowledge, we propose the first poster generation dataset that includes visual attribute annotations for over 76k posters. Qualitative and quantitative outcomes from user studies and experiments substantiate the efficacy of our system and the aesthetic superiority of the generated posters compared to other poster generation methods.