Scholarlink

<p>To address the challenges in generating traditional Chinese landscape paintings—such as complex hierarchical composition, difficulty in capturing brushwork details, and insufficient stylistic coherence—this study proposes LFMDiff, a diffusion-based generative framework. The model employs a hierarchical local LoRA mechanism to preserve global composition while enhancing local hierarchical and textural details. It further introduces dynamic rank and dynamic alpha mechanisms to adaptively allocate model capacity between easily learned and complex features. In addition, an adversarial adaptive flow and a multi-scale maximum mean discrepancy constraint are integrated to improve latent space alignment and stylistic consistency across multiple granularities. A dedicated dataset, CTLPD, is also constructed for model training and evaluation. Experimental results demonstrate that LFMDiff achieves superior performance in global composition, brushstroke hierarchy, and fine-grained detail generation, offering new insights into the integration of artificial intelligence and traditional painting styles.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LFMDiff: generation of Chinese traditional landscape paintings based on diffusion model

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月6日

查看全文
  • Zhijie Li,
  • Yue Wang,
  • Changhua Li,
  • Jie Zhang,
  • Shengjun Xu,
  • Yuan Gao

摘要

To address the challenges in generating traditional Chinese landscape paintings—such as complex hierarchical composition, difficulty in capturing brushwork details, and insufficient stylistic coherence—this study proposes LFMDiff, a diffusion-based generative framework. The model employs a hierarchical local LoRA mechanism to preserve global composition while enhancing local hierarchical and textural details. It further introduces dynamic rank and dynamic alpha mechanisms to adaptively allocate model capacity between easily learned and complex features. In addition, an adversarial adaptive flow and a multi-scale maximum mean discrepancy constraint are integrated to improve latent space alignment and stylistic consistency across multiple granularities. A dedicated dataset, CTLPD, is also constructed for model training and evaluation. Experimental results demonstrate that LFMDiff achieves superior performance in global composition, brushstroke hierarchy, and fine-grained detail generation, offering new insights into the integration of artificial intelligence and traditional painting styles.