Parametric Gaussian Human Model: Generalizable Prior for Efficient and Realistic Human Avatar Modeling
By: Cheng Peng , Jingxiang Sun , Yushuo Chen and more
Potential Business Impact:
Creates realistic people for games from one video.
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.
Similar Papers
HGC-Avatar: Hierarchical Gaussian Compression for Streamable Dynamic 3D Avatars
CV and Pattern Recognition
Makes 3D avatars look real and move smoothly.
GaussianGAN: Real-Time Photorealistic controllable Human Avatars
CV and Pattern Recognition
Makes digital people look real and move smoothly.
FMGS-Avatar: Mesh-Guided 2D Gaussian Splatting with Foundation Model Priors for 3D Monocular Avatar Reconstruction
CV and Pattern Recognition
Makes 3D people from one video.