arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.06606cs.CV
收藏

MaterialFusion:基于扩散模型的高质量零样本可控材质迁移

MaterialFusion: High-Quality, Zero-Shot, and Controllable Material Transfer with Diffusion Models

HSE University(高等经济大学);AIRI;Skolkovo Institute of Science and Technology(斯科尔科沃科学技术学院)

Kamil Garifullin, Maxim Nikolaev, Andrey Kuznetsov, Aibek Alanov

AI总结:

MaterialFusion是基于扩散模型的新型材质迁移框架,可零样本实现高质量可控材质迁移,在质量、可控性等方面优于现有方法,代码公开。

摘要:

操控图像中物体的材质外观对增强现实、虚拟原型设计和数字内容创作等应用至关重要。我们提出MaterialFusion,这一用于高质量材质迁移的新型框架,允许用户调整材质应用程度,在新材料属性与物体原始特征间实现最优平衡。MaterialFusion通过保持背景一致性并缓解边界伪影,将修改后的物体无缝融入场景。为全面评估该方法,我们编制了真实世界材质迁移示例数据集并开展复杂对比分析。通过综合定量评估与用户研究,我们证明MaterialFusion在质量、用户可控性和背景保留方面显著优于现有方法。代码可在指定URL获取。

英文摘要:

Manipulating the material appearance of objects in images is critical for applications like augmented reality, virtual prototyping, and digital content creation. We present MaterialFusion, a novel framework for high-quality material transfer that allows users to adjust the degree of material application, achieving an optimal balance between new material properties and the object's original features. MaterialFusion seamlessly integrates the modified object into the scene by maintaining background consistency and mitigating boundary artifacts. To thoroughly evaluate our approach, we have compiled a dataset of real-world material transfer examples and conducted complex comparative analyses. Through comprehensive quantitative evaluations and user studies, we demonstrate that MaterialFusion significantly outperforms existing methods in terms of quality, user control, and background preservation. Code is available at https://github.com/ControlGenAI/MaterialFusion.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑