Color Me Correctly: Bridging Perceptual Color Spaces and Text Embeddings for Improved Diffusion Generation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Tsai, Sung-Lin, Huang, Bo-Lun, Shen, Yu Ting, Yeo, Cheng Yu, Tseng, Chiang, Ruan, Bo-Kai, Lien, Wen-Sheng, Shuai, Hong-Han
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912584740896768
author Tsai, Sung-Lin
Huang, Bo-Lun
Shen, Yu Ting
Yeo, Cheng Yu
Tseng, Chiang
Ruan, Bo-Kai
Lien, Wen-Sheng
Shuai, Hong-Han
author_facet Tsai, Sung-Lin
Huang, Bo-Lun
Shen, Yu Ting
Yeo, Cheng Yu
Tseng, Chiang
Ruan, Bo-Kai
Lien, Wen-Sheng
Shuai, Hong-Han
contents Accurate color alignment in text-to-image (T2I) generation is critical for applications such as fashion, product visualization, and interior design, yet current diffusion models struggle with nuanced and compound color terms (e.g., Tiffany blue, lime green, hot pink), often producing images that are misaligned with human intent. Existing approaches rely on cross-attention manipulation, reference images, or fine-tuning but fail to systematically resolve ambiguous color descriptions. To precisely render colors under prompt ambiguity, we propose a training-free framework that enhances color fidelity by leveraging a large language model (LLM) to disambiguate color-related prompts and guiding color blending operations directly in the text embedding space. Our method first employs a large language model (LLM) to resolve ambiguous color terms in the text prompt, and then refines the text embeddings based on the spatial relationships of the resulting color terms in the CIELAB color space. Unlike prior methods, our approach improves color accuracy without requiring additional training or external reference images. Experimental results demonstrate that our framework improves color alignment without compromising image quality, bridging the gap between text semantics and visual generation.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10058
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Color Me Correctly: Bridging Perceptual Color Spaces and Text Embeddings for Improved Diffusion Generation
Tsai, Sung-Lin
Huang, Bo-Lun
Shen, Yu Ting
Yeo, Cheng Yu
Tseng, Chiang
Ruan, Bo-Kai
Lien, Wen-Sheng
Shuai, Hong-Han
Computer Vision and Pattern Recognition
Accurate color alignment in text-to-image (T2I) generation is critical for applications such as fashion, product visualization, and interior design, yet current diffusion models struggle with nuanced and compound color terms (e.g., Tiffany blue, lime green, hot pink), often producing images that are misaligned with human intent. Existing approaches rely on cross-attention manipulation, reference images, or fine-tuning but fail to systematically resolve ambiguous color descriptions. To precisely render colors under prompt ambiguity, we propose a training-free framework that enhances color fidelity by leveraging a large language model (LLM) to disambiguate color-related prompts and guiding color blending operations directly in the text embedding space. Our method first employs a large language model (LLM) to resolve ambiguous color terms in the text prompt, and then refines the text embeddings based on the spatial relationships of the resulting color terms in the CIELAB color space. Unlike prior methods, our approach improves color accuracy without requiring additional training or external reference images. Experimental results demonstrate that our framework improves color alignment without compromising image quality, bridging the gap between text semantics and visual generation.
title Color Me Correctly: Bridging Perceptual Color Spaces and Text Embeddings for Improved Diffusion Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.10058