AI RESEARCH
ReorgGS: Equivalent Distribution Reorganization for 3D Gaussian Splatting
arXiv CS.CV
•
ArXi:2605.08739v1 Announce Type: new A converged 3D Gaussian Splatting (3DGS) model may approximate the target scene while remaining poorly parameterized for further optimization. We identify this failure mode as \emph{parameterization degeneration}: high-opacity floaters attenuate gradients to true surfaces through alpha compositing, and redundant overlapping clusters create strongly coupled parameter blocks with nearly collinear Jacobian responses. These effects explain why continued optimization can plateau even when the model still contains removable artifacts.