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

arXiv: https://arxiv.org/abs/2508.06968 · HTML: https://mlsanity.com/paper/2508.06968/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 41 references that are in the map.
- First posted on arXiv 9 Aug 2025.

## Builds on (most cited first, 17 of 17)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://mlsanity.com/paper/2401.10891/) (2024)
- [Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields](https://mlsanity.com/paper/2304.06706/) (2023)
- [MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision](https://mlsanity.com/paper/2410.19115/) (2024)
- [3D Gaussian Splatting as Markov Chain Monte Carlo](https://mlsanity.com/paper/2404.09591/) (2024)
- [3DGUT: Enabling Distorted Cameras and Secondary Rays in Gaussian Splatting](https://mlsanity.com/paper/2412.12507/) (2024)
- [SMERF: Streamable Memory Efficient Radiance Fields for Real-Time Large-Scene Exploration](https://mlsanity.com/paper/2312.07541/) (2023)
- [UniK3D: Universal Camera Monocular 3D Estimation](https://mlsanity.com/paper/2503.16591/) (2025)
- [The promises and pitfalls of Stochastic Gradient Langevin Dynamics](https://mlsanity.com/paper/1811.10072/) (2018)
- [On the Error Analysis of 3D Gaussian Splatting and an Optimal Projection Strategy](https://mlsanity.com/paper/2402.00752/) (2024)
- [FisheyeDistanceNet: Self-Supervised Scale-Aware Distance Estimation using Monocular Fisheye Camera for Autonomous Driving](https://mlsanity.com/paper/1910.04076/) (2019)
- [UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a Generic Framework for Handling Common Camera Distortion Models](https://mlsanity.com/paper/2007.06676/) (2020)
- [Fisheye-GS: Lightweight and Extensible Gaussian Splatting Module for Fisheye Cameras](https://mlsanity.com/paper/2409.04751/) (2024)
- [Self-Calibrating Gaussian Splatting for Large Field of View Reconstruction](https://mlsanity.com/paper/2502.09563/) (2025)
- [FisheyeDepth: A Real Scale Self-Supervised Depth Estimation Model for Fisheye Camera](https://mlsanity.com/paper/2409.15054/) (2024)
- [FIORD: A Fisheye Indoor-Outdoor Dataset with LIDAR Ground Truth for 3D Scene Reconstruction and Benchmarking](https://mlsanity.com/paper/2504.01732/) (2025)

## Abstract (arXiv)

We present the first evaluation of 3D Gaussian Splatting methods on real fisheye imagery with fields of view above 180textdegree. Our study evaluates Fisheye-GS citeliao2024fisheyegslightweightextensiblegaussian and 3DGUT citewu20253dgut on indoor and outdoor scenes captured with 200textdegree fisheye cameras, with the aim of assessing the practicality of wide-angle reconstruction under severe distortion. By comparing reconstructions at 200textdegree, 160textdegree, and 120textdegree field-of-view, we show that both methods achieve their best results at 160textdegree, which balances scene coverage with image quality, while distortion at 200textdegree 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) citepiccinelli2025unik3d. This represents the first application of UniK3D to fisheye imagery beyond 200textdegree, 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.