Seal2Real: Prompt Prior Learning on Diffusion Model for Unsupervised Document Seal Data Generation and Realisation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914231338663936 |
|---|---|
| author | Yan, Mingfu Huang, Jiancheng Chen, Shifeng |
| author_facet | Yan, Mingfu Huang, Jiancheng Chen, Shifeng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_00546 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Seal2Real: Prompt Prior Learning on Diffusion Model for Unsupervised Document Seal Data Generation and Realisation Yan, Mingfu Huang, Jiancheng Chen, Shifeng Computer Vision and Pattern Recognition 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. |
| title | Seal2Real: Prompt Prior Learning on Diffusion Model for Unsupervised Document Seal Data Generation and Realisation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2310.00546 |