SketchFaceGS：基于高斯泼溅的实时草图驱动人脸编辑与生成
SketchFaceGS: Real-Time Sketch-Driven Face Editing and Generation with Gaussian Splatting
把 2D 草图作为稀疏、深度歧义的交互条件，用 Transformer 前馈预测 UV 特征图再细化成 3D 高斯头部，为「草图→三维表示」的单次前向生成与自由视角编辑提供了可复用的 UV 空间特征融合与掩码融合思路。
原文依据 the first sketch-driven framework for real-time generation and editing of photorealistic 3D Gaussian head models from 2D sketches
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论三维高斯表示已成为数字头部建模的一种强大范式，能够以实时渲染达到照片级真实感。然而，对三维高斯头部模型进行直观且交互式的创建或编辑仍然具有挑战性。尽管二维草图提供了快速、直观的概念设计的理想交互模态，但草图稀疏、深度歧义且缺乏高频外观线索，使得从笔画推断稠密且几何一致的三维高斯结构十分困难，在实时约束下尤为如此。为应对这些挑战，我们提出 SketchFaceGS，这是首个从二维草图实时生成与编辑照片级真实感三维高斯头部模型的草图驱动框架。本方法采用前馈的由粗到细架构。基于 Transformer 的 UV 特征预测模块首先从输入草图重建粗略但几何一致的 UV 特征图，随后三维 UV 特征增强模块以高频、照片级真实的细节对其进行细化，从而生成高保真三维头部。在编辑方面，我们引入 UV 掩码融合技术并结合逐层特征融合策略，实现精确、实时、自由视角的修改。大量实验表明，SketchFaceGS 在生成保真度与编辑灵活性上均优于现有方法，可在单次前向传播中由草图生成高质量、可编辑的三维头部。
英文原文摘要 1464 字符
3D Gaussian representations have emerged as a powerful paradigm for digital head modeling, achieving photorealistic quality with real-time rendering. However, intuitive and interactive creation or editing of 3D Gaussian head models remains challenging. Although 2D sketches provide an ideal interaction modality for fast, intuitive conceptual design, they are sparse, depth-ambiguous, and lack high-frequency appearance cues, making it difficult to infer dense, geometrically consistent 3D Gaussian structures from strokes--especially under real-time constraints. To address these challenges, we propose SketchFaceGS, the first sketch-driven framework for real-time generation and editing of photorealistic 3D Gaussian head models from 2D sketches. Our method uses a feed-forward, coarse-to-fine architecture. A Transformer-based UV feature-prediction module first reconstructs a coarse but geometrically consistent UV feature map from the input sketch, and then a 3D UV feature enhancement module refines it with high-frequency, photorealistic detail to produce a high-fidelity 3D head. For editing, we introduce a UV Mask Fusion technique combined with a layer-by-layer feature-fusion strategy, enabling precise, real-time, free-viewpoint modifications. Extensive experiments show that SketchFaceGS outperforms existing methods in both generation fidelity and editing flexibility, producing high-quality, editable 3D heads from sketches in a single forward pass.