3D style transfer enables the creation of visually expressive 3D content, enriching the visual appearance of 3D scenes and objects. However, existing VGG- and CLIP-based methods struggle to model multi-view consistency within the model itself, while diffusion-based approaches can capture such consistency but rely on denoising directions, leading to unstable training. To address these limitations, we propose DiffStyle3D, a novel diffusion-based paradigm for 3DGS style transfer that directly optimizes in the latent space. Specifically, we introduce an Attention-Aware Loss that performs style transfer by aligning style features in the self-attention space, while preserving original content through content feature alignment. Inspired by the geometric invariance of 3D stylization, we propose a Geometry-Guided Multi-View Consistency method that integrates geometric information into self-attention to enable cross-view correspondence modeling. Based on geometric information, we additionally construct a geometry-aware mask to prevent redundant optimization in overlapping regions across views, which further improves multi-view consistency. Extensive experiments show that DiffStyle3D outperforms state-of-the-art methods, achieving higher stylization quality and visual realism.

翻译：三维风格迁移能够创建视觉表现力丰富的三维内容，从而增强三维场景与物体的视觉外观。然而，现有的基于VGG和CLIP的方法难以在模型内部建模多视角一致性，而基于扩散的方法虽能捕捉此类一致性，却依赖于去噪方向，导致训练不稳定。为应对这些局限性，我们提出了DiffStyle3D，一种用于三维高斯风格化的新型扩散范式，可直接在隐空间中进行优化。具体而言，我们引入了一种注意力感知损失，该损失通过在自注意力空间中对齐风格特征来执行风格迁移，同时通过内容特征对齐保留原始内容。受三维风格化几何不变性的启发，我们提出了一种几何引导的多视角一致性方法，将几何信息整合到自注意力中，以实现跨视角对应关系建模。基于几何信息，我们还构建了一个几何感知掩码，以防止跨视角重叠区域的冗余优化，从而进一步提升多视角一致性。大量实验表明，DiffStyle3D在风格化质量和视觉真实感方面均优于现有先进方法。