PiG-Avatar decouples Gaussian avatar geometry from body-template surfaces by anchoring Gaussians in a neural-field-governed volumetric canonical space and using barycentric transport for kinematics, yielding SOTA rendering on complex-clothing benchmarks.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
DiffRGD is a plug-and-play inference-time guidance method that casts each diffusion sampling step as constrained optimization on a spherical manifold and solves it with Riemannian gradient descent to preserve the Gaussian latent structure.
citing papers explorer
-
PiG-Avatar: Hierarchical Neural-Field-Guided Gaussian Avatars
PiG-Avatar decouples Gaussian avatar geometry from body-template surfaces by anchoring Gaussians in a neural-field-governed volumetric canonical space and using barycentric transport for kinematics, yielding SOTA rendering on complex-clothing benchmarks.
-
DiffRGD: An Inference-Time Diffusion Guidance Through Riemannian Gradient Descent
DiffRGD is a plug-and-play inference-time guidance method that casts each diffusion sampling step as constrained optimization on a spherical manifold and solves it with Riemannian gradient descent to preserve the Gaussian latent structure.