Predictive Photometric Uncertainty in Gaussian Splatting for Novel View Synthesis

2026-03-24

AI Summary

This work addresses the lack of reliable photometric uncertainty estimation in 3D Gaussian splatting for novel view synthesis, which hinders its trustworthy deployment in safety-critical applications. The authors propose a lightweight, plug-and-play post-processing framework that introduces per-Gaussian uncertainty channels without modifying the original representation. By modeling reconstruction residuals through Bayesian-regularized linear least squares optimization, the method enables pixel-wise, view-dependent predictive uncertainty quantification. While preserving rendering fidelity, the approach significantly enhances performance in three downstream tasks: active viewpoint selection, pose-invariant scene change detection, and anomaly detection. This advancement represents a crucial step toward enabling trustworthy spatial perception with Gaussian splatting.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Computer Vision: 3D Computer Vision

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Security and Privacy: Data transparency and provenance
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Recent advances in 3D Gaussian Splatting have enabled impressive photorealistic novel view synthesis. However, to transition from a pure rendering engine to a reliable spatial map for autonomous agents and safety-critical applications, knowing where the representation is uncertain is as important as the rendering fidelity itself. We bridge this critical gap by introducing a lightweight, plug-and-play framework for pixel-wise, view-dependent predictive uncertainty estimation. Our post-hoc method formulates uncertainty as a Bayesian-regularized linear least-squares optimization over reconstruction residuals. This architecture-agnostic approach extracts a per-primitive uncertainty channel without modifying the underlying scene representation or degrading baseline visual fidelity. Crucially, we demonstrate that providing this actionable reliability signal successfully translates 3D Gaussian splatting into a trustworthy spatial map, further improving state-of-the-art performance across three critical downstream perception tasks: active view selection, pose-agnostic scene change detection, and pose-agnostic anomaly detection.

Problem

• Predictive Uncertainty
• Gaussian Splatting
• Novel View Synthesis
• Photometric Uncertainty
• Spatial Mapping

Innovation

• Predictive Uncertainty
• Gaussian Splatting
• Novel View Synthesis
• Bayesian Regularization
• Spatial Mapping