3D Gaussian Splatting with Fisheye Images: Field of View Analysis and Depth-Based Initialization

TL;DR

This paper evaluates 3D Gaussian Splatting on real fisheye images with over 180° FOV, introduces a depth-based initialization method using UniK3D for wide-angle scenes, and benchmarks reconstruction quality under severe distortion.

Contribution

It is the first to assess 3D Gaussian Splatting on fisheye images above 200° FOV and introduces a novel depth-based initialization approach with UniK3D for improved reconstruction.

Findings

Optimal FOV for reconstruction is around 160° due to distortion effects.

UniK3D can produce accurate 3D reconstructions without training on fisheye data.

Wide-angle fisheye images enable feasible 3D reconstruction with Gaussian Splatting.

Abstract

We present the first evaluation of 3D Gaussian Splatting methods on real fisheye imagery with fields of view above 180\textdegree{}. Our study evaluates Fisheye-GS \cite{liao2024fisheyegslightweightextensiblegaussian} and 3DGUT \cite{wu20253dgut} on indoor and outdoor scenes captured with 200\textdegree{} fisheye cameras, with the aim of assessing the practicality of wide-angle reconstruction under severe distortion. By comparing reconstructions at 200\textdegree{}, 160\textdegree{}, and 120\textdegree{} field-of-view, we show that both methods achieve their best results at 160\textdegree{}, which balances scene coverage with image quality, while distortion at 200\textdegree{} degrades performance. To address the common failure of Structure-from-Motion (SfM) initialization at such wide angles, we introduce a depth-based alternative using UniK3D (Universal Camera Monocular 3D Estimation) \cite{piccinelli2025unik3d}. This represents the first application of UniK3D to fisheye imagery beyond 200\textdegree{}, despite the model not being trained on such data. With the number of predicted points controlled to match SfM for fairness, UniK3D produces geometrically accurate reconstructions that rival or surpass SfM, even in challenging scenes with fog, glare, or open sky. These results demonstrate the feasibility of fisheye-based 3D Gaussian Splatting and provides a benchmark for future research on wide-angle reconstruction from sparse and distorted inputs.