ERGO: Excess-Risk-Guided Optimization for High-Fidelity Monocular 3D Gaussian Splatting

2026-02-10

AI Summary

Monocular 3D reconstruction often suffers from missing geometry and texture due to occlusions, while auxiliary views can introduce geometric inconsistencies and texture misalignment. To address these challenges, this work proposes ERGO, a novel framework that, for the first time, integrates excess risk decomposition into 3D Gaussian splatting optimization. By dynamically estimating the optimizable excess risk and the irreducible Bayesian error for each view, ERGO establishes a view-adaptive loss weighting mechanism that jointly optimizes geometry- and texture-aware objectives in a global-local collaborative manner. Extensive experiments on Google Scanned Objects and OmniObject3D demonstrate that ERGO significantly outperforms existing methods, achieving superior geometric fidelity and texture quality in monocular 3D reconstruction.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Non-convex Optimization

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Generating 3D content from a single image remains a fundamentally challenging and ill-posed problem due to the inherent absence of geometric and textural information in occluded regions. While state-of-the-art generative models can synthesize auxiliary views to provide additional supervision, these views inevitably contain geometric inconsistencies and textural misalignments that propagate and amplify artifacts during 3D reconstruction. To effectively harness these imperfect supervisory signals, we propose an adaptive optimization framework guided by excess risk decomposition, termed ERGO. Specifically, ERGO decomposes the optimization losses in 3D Gaussian splatting into two components, i.e., excess risk that quantifies the suboptimality gap between current and optimal parameters, and Bayes error that models the irreducible noise inherent in synthesized views. This decomposition enables ERGO to dynamically estimate the view-specific excess risk and adaptively adjust loss weights during optimization. Furthermore, we introduce geometry-aware and texture-aware objectives that complement the excess-risk-derived weighting mechanism, establishing a synergistic global-local optimization paradigm. Consequently, ERGO demonstrates robustness against supervision noise while consistently enhancing both geometric fidelity and textural quality of the reconstructed 3D content. Extensive experiments on the Google Scanned Objects dataset and the OmniObject3D dataset demonstrate the superiority of ERGO over existing state-of-the-art methods.

Problem

• monocular 3D reconstruction
• geometric inconsistency
• textural misalignment
• occlusion
• supervision noise

Innovation

• Excess Risk
• 3D Gaussian Splatting
• Monocular 3D Reconstruction
• Adaptive Optimization
• Geometry-Aware Loss