UniGlyph: Unified Segmentation-Conditioned Diffusion for Precise Visual Text Synthesis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913921970995200 |
|---|---|
| author | Wang, Yuanrui Han, Cong Li, Yafei Jin, Zhipeng Li, Xiawei Du, SiNan Tao, Wen Yang, Yi Li, Shuanglong Yuan, Chun Lin, Liu |
| author_facet | Wang, Yuanrui Han, Cong Li, Yafei Jin, Zhipeng Li, Xiawei Du, SiNan Tao, Wen Yang, Yi Li, Shuanglong Yuan, Chun Lin, Liu |
| contents | Text-to-image generation has greatly advanced content creation, yet accurately rendering visual text remains a key challenge due to blurred glyphs, semantic drift, and limited style control. Existing methods often rely on pre-rendered glyph images as conditions, but these struggle to retain original font styles and color cues, necessitating complex multi-branch designs that increase model overhead and reduce flexibility. To address these issues, we propose a segmentation-guided framework that uses pixel-level visual text masks -- rich in glyph shape, color, and spatial detail -- as unified conditional inputs. Our method introduces two core components: (1) a fine-tuned bilingual segmentation model for precise text mask extraction, and (2) a streamlined diffusion model augmented with adaptive glyph conditioning and a region-specific loss to preserve textual fidelity in both content and style. Our approach achieves state-of-the-art performance on the AnyText benchmark, significantly surpassing prior methods in both Chinese and English settings. To enable more rigorous evaluation, we also introduce two new benchmarks: GlyphMM-benchmark for testing layout and glyph consistency in complex typesetting, and MiniText-benchmark for assessing generation quality in small-scale text regions. Experimental results show that our model outperforms existing methods by a large margin in both scenarios, particularly excelling at small text rendering and complex layout preservation, validating its strong generalization and deployment readiness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00992 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | UniGlyph: Unified Segmentation-Conditioned Diffusion for Precise Visual Text Synthesis Wang, Yuanrui Han, Cong Li, Yafei Jin, Zhipeng Li, Xiawei Du, SiNan Tao, Wen Yang, Yi Li, Shuanglong Yuan, Chun Lin, Liu Computer Vision and Pattern Recognition Text-to-image generation has greatly advanced content creation, yet accurately rendering visual text remains a key challenge due to blurred glyphs, semantic drift, and limited style control. Existing methods often rely on pre-rendered glyph images as conditions, but these struggle to retain original font styles and color cues, necessitating complex multi-branch designs that increase model overhead and reduce flexibility. To address these issues, we propose a segmentation-guided framework that uses pixel-level visual text masks -- rich in glyph shape, color, and spatial detail -- as unified conditional inputs. Our method introduces two core components: (1) a fine-tuned bilingual segmentation model for precise text mask extraction, and (2) a streamlined diffusion model augmented with adaptive glyph conditioning and a region-specific loss to preserve textual fidelity in both content and style. Our approach achieves state-of-the-art performance on the AnyText benchmark, significantly surpassing prior methods in both Chinese and English settings. To enable more rigorous evaluation, we also introduce two new benchmarks: GlyphMM-benchmark for testing layout and glyph consistency in complex typesetting, and MiniText-benchmark for assessing generation quality in small-scale text regions. Experimental results show that our model outperforms existing methods by a large margin in both scenarios, particularly excelling at small text rendering and complex layout preservation, validating its strong generalization and deployment readiness. |
| title | UniGlyph: Unified Segmentation-Conditioned Diffusion for Precise Visual Text Synthesis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.00992 |