Multi-StyleGS：使用多种风格对 Gaussian Splatting 进行风格化
Multi-StyleGS: Stylizing Gaussian Splatting with Multiple Styles
AI总结：
本文提出 Multi-StyleGS，通过二分匹配、语义风格损失、局部—全局特征匹配及分割正则化，实现多风格三维 GS 的自动局部迁移与灵活编辑，并提升内存效率、纹理和色彩表现。
摘要：
近年来，出于创意目的，将给定三维场景风格化以使其与参考图像的艺术风格保持一致的需求不断增长。尽管三维 Gaussian Splatting（GS）已成为一种用于真实感三维场景建模的有前景且高效的方法，但在使其适应三维 GS 风格化以通过自动局部风格迁移或人工指定来匹配多种风格，同时在风格化训练中保持内存效率方面，仍然存在挑战。本文提出一种名为 Multi-StyleGS 的新型三维 GS 风格化方案以应对这些挑战。具体而言，我们采用二分匹配机制自动识别风格图像与渲染图像局部区域之间的对应关系。为促进局部风格迁移，我们提出一种新的语义风格损失函数，该函数利用分割网络为场景中的不同对象施加不同风格，并提出局部—全局特征匹配以增强多视角一致性。此外，该技术能够实现内存高效的训练、更丰富的纹理细节和更好的色彩匹配。为更好地为每个 Gaussian 分配鲁棒的语义标签，我们提出若干技术来正则化分割网络。综合实验表明，我们的方法在生成合理的风格化结果和提供灵活编辑方面优于现有方法。
英文摘要：
In recent years, there has been a growing demand to stylize a given 3D scene to align with the artistic style of reference images for creative purposes. While 3D Gaussian Splatting(GS) has emerged as a promising and efficient method for realistic 3D scene modeling, there remains a challenge in adapting it to stylize 3D GS to match with multiple styles through automatic local style transfer or manual designation, while maintaining memory efficiency for stylization training. In this paper, we introduce a novel 3D GS stylization solution termed Multi-StyleGS to tackle these challenges. In particular, we employ a bipartite matching mechanism to au tomatically identify correspondences between the style images and the local regions of the rendered images. To facilitate local style transfer, we introduce a novel semantic style loss function that employs a segmentation network to apply distinct styles to various objects of the scene and propose a local-global feature matching to enhance the multi-view consistency. Furthermore, this technique can achieve memory efficient training, more texture details and better color match. To better assign a robust semantic label to each Gaussian, we propose several techniques to regularize the segmentation network. As demonstrated by our comprehensive experiments, our approach outperforms existing ones in producing plausible stylization results and offering flexible editing.
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。
