StyleTextGen: Style-Conditioned Multilingual Scene Text Generation

• URL: http://arxiv.org/abs/2605.14708v1
• Date: Thu, 14 May 2026 11:24:44 GMT
• Title: StyleTextGen: Style-Conditioned Multilingual Scene Text Generation
• Abstract summary: StyleTextGen is a novel framework that learns to perceive and replicate visual text styles across different languages and writing systems.<n>Our approach features three key contributions: First, we introduce a dual-branch style encoder dedicated to style modeling.<n>Second, we design a text style consistency loss that enhances style coherence.<n>Third, we develop a mask-guided inference strategy that ensures precise style alignment between generated and reference text.
• Score: 19.235225130538794
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Style-conditioned scene text generation faces unique challenges in extracting precise text styles from complex backgrounds and maintaining fine-grained style consistency across characters, especially for multilingual scripts. We propose StyleTextGen, a novel framework that learns to perceive and replicate visual text styles across different languages and writing systems. Our approach features three key contributions: First, we introduce a dual-branch style encoder dedicated to style modeling, yielding robust multilingual text style representations in complex real-world scenes. Second, we design a text style consistency loss that enhances style coherence and improves overall visual quality. Third, we develop a mask-guided inference strategy that ensures precise style alignment between generated and reference text. To facilitate systematic evaluation, we construct StyleText-CE, a bilingual scene text style benchmark covering both monolingual and cross-lingual settings. Extensive experiments demonstrate that StyleTextGen significantly outperforms existing methods in style consistency and cross-lingual generalization, establishing new state-of-the-art performance in multilingual style-conditioned text generation.

Related papers

• Enhanced Generative Structure Prior for Chinese Text Image Super-resolution [101.66745917380837]
  We introduce a text image framework designed to restore the precise strokes of low-resolution (LR) Chinese characters.<n>Our framework incorporates this structure prior within a StyleGAN model.<n>Our code and pre-trained models will be available at https://github.com/csi2016/MARCONetPlus.
  arXiv  Detail & Related papers  (2025-08-11T01:34:45Z)
• StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models [10.685779311280266]
  StyleBlend is a method designed to learn and apply style representations from a limited set of reference images.<n>Our approach decomposes style into two components, composition and texture, each learned through different strategies.
  arXiv  Detail & Related papers  (2025-02-13T08:26:54Z)
• StAyaL | Multilingual Style Transfer [0.0]
  We show that by leveraging only 100 lines of text, an individuals unique style can be captured as a high-dimensional embedding.<n>This methodology breaks down the language barrier by transferring the style of a speaker between languages.<n>The proposed approach is shown to be topic-agnostic, with test accuracy and F1 scores of 74.9% and 0.75, respectively.
  arXiv  Detail & Related papers  (2025-01-20T18:13:18Z)
• Towards Visual Text Design Transfer Across Languages [49.78504488452978]
  We introduce a novel task of Multimodal Style Translation (MuST-Bench) MuST-Bench is a benchmark designed to evaluate the ability of visual text generation models to perform translation across different writing systems. In response, we introduce SIGIL, a framework for multimodal style translation that eliminates the need for style descriptions.
  arXiv  Detail & Related papers  (2024-10-24T15:15:01Z)
• TextMastero: Mastering High-Quality Scene Text Editing in Diverse Languages and Styles [12.182588762414058]
  Scene text editing aims to modify texts on images while maintaining the style of newly generated text similar to the original. Recent works leverage diffusion models, showing improved results, yet still face challenges. We present emphTextMastero - a carefully designed multilingual scene text editing architecture based on latent diffusion models (LDMs)
  arXiv  Detail & Related papers  (2024-08-20T08:06:09Z)
• StyleMamba : State Space Model for Efficient Text-driven Image Style Transfer [9.010012117838725]
  StyleMamba is an efficient image style transfer framework that translates text prompts into corresponding visual styles. Existing text-guided stylization requires hundreds of training iterations and takes a lot of computing resources.
  arXiv  Detail & Related papers  (2024-05-08T12:57:53Z)
• Style Aligned Image Generation via Shared Attention [61.121465570763085]
  We introduce StyleAligned, a technique designed to establish style alignment among a series of generated images. By employing minimal attention sharing' during the diffusion process, our method maintains style consistency across images within T2I models. Our method's evaluation across diverse styles and text prompts demonstrates high-quality and fidelity.
  arXiv  Detail & Related papers  (2023-12-04T18:55:35Z)
• StyleCrafter: Enhancing Stylized Text-to-Video Generation with Style Adapter [78.75422651890776]
  StyleCrafter is a generic method that enhances pre-trained T2V models with a style control adapter. To promote content-style disentanglement, we remove style descriptions from the text prompt and extract style information solely from the reference image. StyleCrafter efficiently generates high-quality stylized videos that align with the content of the texts and resemble the style of the reference images.
  arXiv  Detail & Related papers  (2023-12-01T03:53:21Z)
• Stylized Data-to-Text Generation: A Case Study in the E-Commerce Domain [53.22419717434372]
  We propose a new task, namely stylized data-to-text generation, whose aim is to generate coherent text according to a specific style. This task is non-trivial, due to three challenges: the logic of the generated text, unstructured style reference, and biased training samples. We propose a novel stylized data-to-text generation model, named StyleD2T, comprising three components: logic planning-enhanced data embedding, mask-based style embedding, and unbiased stylized text generation.
  arXiv  Detail & Related papers  (2023-05-05T03:02:41Z)
• Handwriting Transformers [98.3964093654716]
  We propose a transformer-based styled handwritten text image generation approach, HWT, that strives to learn both style-content entanglement and global and local writing style patterns. The proposed HWT captures the long and short range relationships within the style examples through a self-attention mechanism. Our proposed HWT generates realistic styled handwritten text images and significantly outperforms the state-of-the-art demonstrated.
  arXiv  Detail & Related papers  (2021-04-08T17:59:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.