Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis

TL;DR

This paper introduces SCGaussian, a novel Gaussian Splatting method that enforces 3D scene structure consistency using matching priors, significantly improving sparse input view synthesis in large scenes with state-of-the-art efficiency.

Contribution

The paper proposes a hybrid Gaussian representation and a structure optimization method that enhances 3D scene consistency in Gaussian Splatting for novel view synthesis.

Findings

Achieves state-of-the-art performance on various scene types.

Effectively handles sparse input views and large scenes.

Demonstrates high efficiency in rendering.

Abstract

Despite the substantial progress of novel view synthesis, existing methods, either based on the Neural Radiance Fields (NeRF) or more recently 3D Gaussian Splatting (3DGS), suffer significant degradation when the input becomes sparse. Numerous efforts have been introduced to alleviate this problem, but they still struggle to synthesize satisfactory results efficiently, especially in the large scene. In this paper, we propose SCGaussian, a Structure Consistent Gaussian Splatting method using matching priors to learn 3D consistent scene structure. Considering the high interdependence of Gaussian attributes, we optimize the scene structure in two folds: rendering geometry and, more importantly, the position of Gaussian primitives, which is hard to be directly constrained in the vanilla 3DGS due to the non-structure property. To achieve this, we present a hybrid Gaussian representation. Besides the ordinary non-structure Gaussian primitives, our model also consists of ray-based Gaussian primitives that are bound to matching rays and whose optimization of their positions is restricted along the ray. Thus, we can utilize the matching correspondence to directly enforce the position of these Gaussian primitives to converge to the surface points where rays intersect. Extensive experiments on forward-facing, surrounding, and complex large scenes show the effectiveness of our approach with state-of-the-art performance and high efficiency. Code is available at https://github.com/prstrive/SCGaussian.