EasySplat: View-Adaptive Learning makes 3D Gaussian Splatting Easy

TL;DR

EasySplat introduces a view-adaptive learning framework that improves 3D Gaussian Splatting by replacing SfM with a view similarity-based point cloud initialization and an adaptive densification strategy, resulting in better scene modeling.

Contribution

The paper presents a novel initialization and densification approach for 3D Gaussian Splatting that enhances efficiency and accuracy without relying on traditional SfM methods.

Findings

Outperforms state-of-the-art in novel view synthesis

Provides high-quality 3D scene reconstructions

Achieves efficient and robust scene initialization

Abstract

3D Gaussian Splatting (3DGS) techniques have achieved satisfactory 3D scene representation. Despite their impressive performance, they confront challenges due to the limitation of structure-from-motion (SfM) methods on acquiring accurate scene initialization, or the inefficiency of densification strategy. In this paper, we introduce a novel framework EasySplat to achieve high-quality 3DGS modeling. Instead of using SfM for scene initialization, we employ a novel method to release the power of large-scale pointmap approaches. Specifically, we propose an efficient grouping strategy based on view similarity, and use robust pointmap priors to obtain high-quality point clouds and camera poses for 3D scene initialization. After obtaining a reliable scene structure, we propose a novel densification approach that adaptively splits Gaussian primitives based on the average shape of neighboring Gaussian ellipsoids, utilizing KNN scheme. In this way, the proposed method tackles the limitation on initialization and optimization, leading to an efficient and accurate 3DGS modeling. Extensive experiments demonstrate that EasySplat outperforms the current state-of-the-art (SOTA) in handling novel view synthesis.