When 3D Gaussian Splatting Recovers Real Surfaces
cs.LG, cs.CV
Submitted: 2026-08-30
Updated: 2026-08-30
Comments: Accepted at ECCV 2026. 30 pages, 3 figures, 1 table; includes supplementary material
License: http://creativecommons.org/licenses/by/4.0/
The gist: When does 3D Gaussian Splatting (3DGS) recover the true scene surface rather than just overfitting view-dependent appearance? We answer this by developing a mathematical framework based on a
Terminology
Abstract
When does 3D Gaussian Splatting (3DGS) recover the true scene surface rather than just overfitting view-dependent appearance? We answer this by developing a mathematical framework based on a first-hit rendering abstraction that cleanly isolates geometry from appearance. We prove that geometric misalignment forcefully converts spatial textures into high-frequency angular signals via parallax. This establishes a strict identifiability window: if angular capacity is bounded, surface-consistent solutions are mathematically preferred; if unrestricted, the same images can be perfectly explained by an incorrect, opaque billboard geometry. Experiments on synthetic stress tests confirm this prediction, showing billboard failures emerge precisely at high angular capacities. Conversely, in the real-world datasets we evaluate under standard capture protocols, reconstructions remain surface-consistent even at high SH degrees, which is consistent with the prediction that rich spatial texture can push billboard solutions outside the tested angular-capacity range.
Sources
- SolidGS: Consolidating Gaussian Surfel Splatting for Sparse-View Surface Reconstruction
- Gaussian-Augmented Physics Simulation and System Identification with Complex Colliders
- GausSurf: Geometry-Guided 3D Gaussian Splatting for Surface Reconstruction
- NeRF++: Analyzing and Improving Neural Radiance Fields
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks