A VAE with transformer attention and Hessian-guided sampling is proposed to extrapolate 3D Gaussian Splatting scenes beyond their training resolution, claiming 0.015s inference and improved Chamfer distance and CenseoQoE scores.
Su- pergaussian: Repurposing video models for 3d super reso- lution
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.GR 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
GaussianVAE: Adaptive Learning Dynamics of 3D Gaussians for High-Fidelity Super-Resolution
A VAE with transformer attention and Hessian-guided sampling is proposed to extrapolate 3D Gaussian Splatting scenes beyond their training resolution, claiming 0.015s inference and improved Chamfer distance and CenseoQoE scores.