PoemColor framework revives ancient paintings with poetic semantics

Researchers have developed PoemColor, a novel framework designed to restore the original colors of ancient paintings by aligning poetic cultural semantics with visual restoration. This method addresses the challenges of historical ambiguity and modern semantic bias in colorization. PoemColor utilizes a Poetic Painting Projector to convert poetic context into classical color conditions and employs Structure-Aware Semantic Attention to control the injection of poetic color semantics into diffusion models. The framework was evaluated using a hybrid dataset and demonstrated superior performance in delivering historically authentic and semantically rich colorizations. AI

IMPACT This research could lead to more accurate and culturally sensitive AI-driven restoration of historical art.

RANK_REASON The item is an academic paper detailing a new framework for image colorization. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• PoemColor
• Poetic Painting Projector
• SASA
• ScienceCast
• Structure-Aware Semantic Attention
• Wikimedia category

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →