UGOD: Uncertainty-Guided Opacity and Dropout for Sparse-View 3D Gaussian Splatting

2026-09-30

AI Summary

This study addresses the overfitting issue in sparse-view 3D Gaussian Splatting, where insufficient geometric constraints lead to significant degradation in novel view synthesis quality. To mitigate this, we propose the UGOD framework, which estimates the uncertainty of Gaussian primitives and dynamically modulates their rendering contributions to suppress unreliable ones. The core design incorporates a differentiable opacity modulation mechanism and a decoupled soft dropout branch to effectively prevent uncertainty prediction collapse. Furthermore, precise control is achieved through a lightweight uncertainty head, attribute- and view-conditioned directional scoring, and continuous retention masks. Extensive experiments on the Mip-NeRF 360 and LLFF datasets demonstrate that our method substantially improves sparse-view synthesis quality while yielding more compact Gaussian representations.

Abstract

Sparse-view 3D Gaussian Splatting is prone to overfitting because limited observations leave many Gaussian primitives weakly constrained, yet their contributions are still accumulated through alpha blending. Without uncertainty estimation, the renderer cannot distinguish unreliable primitives from well-constrained ones, allowing their erroneous contributions to corrupt novel-view synthesis. We introduce UGOD, an uncertainty-guided framework that estimates a view-dependent uncertainty score for each Gaussian and uses it to regulate its rendering contribution. A lightweight uncertainty head conditioned on Gaussian attributes and viewing direction predicts this score, which then drives a differentiable opacity-modulation mechanism that attenuates high-uncertainty primitives before compositing. During training, a detached soft-dropout branch applies an uncertainty-controlled continuous keep mask to discourage the model from relying on poorly constrained Gaussians and thereby reduce overfitting. Crucially, detaching the uncertainty score prevents gradients from this stochastic regulariser from biasing or collapsing the uncertainty prediction. Experiments on Mip-NeRF~360 and LLFF show that UGOD improves sparse-view novel-view synthesis while producing more compact Gaussian representations than the compared methods. These results demonstrate that Gaussian uncertainty provides an effective rendering-time control for sparse-view reconstruction.

Problem

• Sparse-view 3D Gaussian Splatting
• Overfitting
• Uncertainty estimation
• Novel-view synthesis

Innovation

• Uncertainty Estimation
• 3D Gaussian Splatting
• Sparse-View Reconstruction
• Opacity Modulation
• Soft-Dropout