参数化高斯人体模型：用于高效、逼真人体化身建模的可泛化先验
Parametric Gaussian Human Model: Generalizable Prior for Efficient and Realistic Human Avatar Modeling
AI总结：
针对3DGS人体化身逐对象优化慢、单目稀疏输入泛化差的问题，提出PGHM框架，利用UV对齐身份特征和解耦多头U-Net，在约20分钟内生成高保真可动画化身。
摘要：
照片级真实且可动画化的人体化身是虚拟/增强现实、远程呈现和数字娱乐的关键使能因素。尽管 3D Gaussian Splatting（3DGS，3D高斯泼溅）的最新进展显著提升了渲染质量和效率，但现有方法仍面临根本性挑战，包括耗时的逐对象优化，以及在稀疏单目输入下泛化能力较差。在本文中，我们提出 Parametric Gaussian Human Model（PGHM，参数化高斯人体模型），这是一个可泛化且高效的框架，将人体先验集成到 3DGS 中，用于从单目视频快速、高保真地重建化身。PGHM 引入了两个核心组件：（1）UV 对齐的潜在身份映射，将特定对象的几何和外观紧凑地编码到可学习特征张量中；（2）解耦的 Multi-Head U-Net（多头U-Net），通过条件化解码器分解静态、姿态相关和视角相关组件，从而预测高斯属性。该设计使其在具有挑战性的姿态和视角下具备稳健的渲染质量，同时无需多视角采集或漫长优化时间即可实现高效的对象适配。实验表明，PGHM 比从零开始优化的方法效率显著更高，每个对象仅需约20分钟即可生成视觉质量相当的化身，从而证明了其在真实世界单目化身创建中的实际适用性。
英文摘要：
Photorealistic and animatable human avatars are a key enabler for virtual/augmented reality, telepresence, and digital entertainment. While recent advances in 3D Gaussian Splatting (3DGS) have greatly improved rendering quality and efficiency, existing methods still face fundamental challenges, including time-consuming per-subject optimization and poor generalization under sparse monocular inputs. In this work, we present the Parametric Gaussian Human Model (PGHM), a generalizable and efficient framework that integrates human priors into 3DGS for fast and high-fidelity avatar reconstruction from monocular videos. PGHM introduces two core components: (1) a UV-aligned latent identity map that compactly encodes subject-specific geometry and appearance into a learnable feature tensor; and (2) a disentangled Multi-Head U-Net that predicts Gaussian attributes by decomposing static, pose-dependent, and view-dependent components via conditioned decoders. This design enables robust rendering quality under challenging poses and viewpoints, while allowing efficient subject adaptation without requiring multi-view capture or long optimization time. Experiments show that PGHM is significantly more efficient than optimization-from-scratch methods, requiring only approximately 20 minutes per subject to produce avatars with comparable visual quality, thereby demonstrating its practical applicability for real-world monocular avatar creation.
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。
