SubSplat: High-Resolution Pixel-aligned 3DGS via Sub-pixel Gaussian Reparameterization

2026-07-22

AI Summary

This work addresses the challenge in high-resolution pixel-aligned 3D Gaussian splatting, where achieving fine detail fidelity often comes at prohibitive computational cost: high-resolution inputs drastically increase memory and runtime demands, while low-resolution representations suffer from insufficient Gaussian density and rendering artifacts. To resolve this trade-off, we propose SubSplat, a novel framework that, for the first time, enables direct reconstruction of high-density Gaussian distributions from low-resolution features. Central to our approach is the Sub-Pixel Gaussian Re-parameterization (SPGR) module, which subdivides primary Gaussians into finer primitives to recover structural density, complemented by multi-view feature aggregation to enhance high-frequency details. SubSplat effectively decouples rendering quality from computational overhead, achieving state-of-the-art performance on RealEstate10K and ACID datasets with both high-fidelity rendering and efficient inference.

Abstract

Pixel-aligned Gaussian splatting enables efficient and generalizable novel-view synthesis. However, high-resolution rendering faces a critical trade-off where increasing input resolution improves detail at the expense of quadratically rising network computational cost. Conversely, maintaining low-resolution inputs stabilizes this cost but results in insufficient Gaussian density and artifacts. To address this, we propose SubSplat, which introduces Sub-pixel Gaussian Reparameterizer(SPGR) to subdivide primary Gaussians into fine-grained primitives, restoring structural density directly from low-resolution features. We further enhance the reparameterization quality through feature aggregation, which effectively captures high-frequency details across multiple views. Experiments on RealEstate10K and ACID demonstrate that SubSplat achieves high-fidelity rendering with superior efficiency. Our results validate that the proposed framework successfully resolves the trade-off between reparameterization fidelity and network computational cost inherent in pixel-aligned Gaussian Splatting.

Problem

• Gaussian splatting
• novel-view synthesis
• high-resolution rendering
• computational cost
• pixel-aligned

Innovation

• Sub-pixel Gaussian Reparameterization
• Pixel-aligned 3DGS
• Novel-view synthesis
• Feature aggregation
• High-resolution rendering