TextPixs: Glyph-Conditioned Diffusion with Character-Aware Attention and OCR-Guided Supervision

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gillani, Syeda Anshrah, Baig, Mirza Samad Ahmed, Khan, Osama Ahmed, Shah, Shahid Munir, Mujeeb, Umema, Ali, Maheen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911045543526400
author Gillani, Syeda Anshrah
Baig, Mirza Samad Ahmed
Khan, Osama Ahmed
Shah, Shahid Munir
Mujeeb, Umema
Ali, Maheen
author_facet Gillani, Syeda Anshrah
Baig, Mirza Samad Ahmed
Khan, Osama Ahmed
Shah, Shahid Munir
Mujeeb, Umema
Ali, Maheen
contents The modern text-to-image diffusion models boom has opened a new era in digital content production as it has proven the previously unseen ability to produce photorealistic and stylistically diverse imagery based on the semantics of natural-language descriptions. However, the consistent disadvantage of these models is that they cannot generate readable, meaningful, and correctly spelled text in generated images, which significantly limits the use of practical purposes like advertising, learning, and creative design. This paper introduces a new framework, namely Glyph-Conditioned Diffusion with Character-Aware Attention (GCDA), using which a typical diffusion backbone is extended by three well-designed modules. To begin with, the model has a dual-stream text encoder that encodes both semantic contextual information and explicit glyph representations, resulting in a character-aware representation of the input text that is rich in nature. Second, an attention mechanism that is aware of the character is proposed with a new attention segregation loss that aims to limit the attention distribution of each character independently in order to avoid distortion artifacts. Lastly, GCDA has an OCR-in-the-loop fine-tuning phase, where a full text perceptual loss, directly optimises models to be legible and accurately spell. Large scale experiments to benchmark datasets, such as MARIO-10M and T2I-CompBench, reveal that GCDA sets a new state-of-the-art on all metrics, with better character based metrics on text rendering (Character Error Rate: 0.08 vs 0.21 for the previous best; Word Error Rate: 0.15 vs 0.25), human perception, and comparable image synthesis quality on high-fidelity (FID: 14.3).
format Preprint
id arxiv_https___arxiv_org_abs_2507_06033
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TextPixs: Glyph-Conditioned Diffusion with Character-Aware Attention and OCR-Guided Supervision
Gillani, Syeda Anshrah
Baig, Mirza Samad Ahmed
Khan, Osama Ahmed
Shah, Shahid Munir
Mujeeb, Umema
Ali, Maheen
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
The modern text-to-image diffusion models boom has opened a new era in digital content production as it has proven the previously unseen ability to produce photorealistic and stylistically diverse imagery based on the semantics of natural-language descriptions. However, the consistent disadvantage of these models is that they cannot generate readable, meaningful, and correctly spelled text in generated images, which significantly limits the use of practical purposes like advertising, learning, and creative design. This paper introduces a new framework, namely Glyph-Conditioned Diffusion with Character-Aware Attention (GCDA), using which a typical diffusion backbone is extended by three well-designed modules. To begin with, the model has a dual-stream text encoder that encodes both semantic contextual information and explicit glyph representations, resulting in a character-aware representation of the input text that is rich in nature. Second, an attention mechanism that is aware of the character is proposed with a new attention segregation loss that aims to limit the attention distribution of each character independently in order to avoid distortion artifacts. Lastly, GCDA has an OCR-in-the-loop fine-tuning phase, where a full text perceptual loss, directly optimises models to be legible and accurately spell. Large scale experiments to benchmark datasets, such as MARIO-10M and T2I-CompBench, reveal that GCDA sets a new state-of-the-art on all metrics, with better character based metrics on text rendering (Character Error Rate: 0.08 vs 0.21 for the previous best; Word Error Rate: 0.15 vs 0.25), human perception, and comparable image synthesis quality on high-fidelity (FID: 14.3).
title TextPixs: Glyph-Conditioned Diffusion with Character-Aware Attention and OCR-Guided Supervision
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.06033