Seal2Real: Prompt Prior Learning on Diffusion Model for Unsupervised Document Seal Data Generation and Realisation

TL;DR

Seal2Real is a novel framework that leverages prompt prior learning on diffusion models to generate large-scale, realistic labeled document seal data, addressing data scarcity in seal-related document processing tasks.

Contribution

The paper introduces Seal2Real, a new generative approach that adapts pre-trained diffusion models for unsupervised seal image synthesis, along with a large labeled dataset Seal-DB.

Findings

Seal2Real produces highly realistic synthetic seal images.

Synthetic data improves downstream seal-related task performance.

Seal-DB contains 20,000 labeled seal images for research.

Abstract

Seal-related tasks in document processing-such as seal segmentation, authenticity verification, seal removal, and text recognition under seals-hold substantial commercial importance. However, progress in these areas has been hindered by the scarcity of labeled document seal datasets, which are essential for supervised learning. To address this limitation, we propose Seal2Real, a novel generative framework designed to synthesize large-scale labeled document seal data. As part of this work, we also present Seal-DB, a comprehensive dataset containing 20,000 labeled images to support seal-related research. Seal2Real introduces a prompt prior learning architecture built upon a pre-trained Stable Diffusion model, effectively transferring its generative capability to the unsupervised domain of seal image synthesis. By producing highly realistic synthetic seal images, Seal2Real significantly enhances the performance of downstream seal-related tasks on real-world data. Experimental evaluations on the Seal-DB dataset demonstrate the effectiveness and practical value of the proposed framework. The dataset is available at https://github.com/liuyifan6613/DocBank-Document-Enhancement-Dataset.