Self-Calibrating Gaussian Splatting for Large Field of View Reconstruction

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Deng, Youming, Xian, Wenqi, Yang, Guandao, Guibas, Leonidas, Wetzstein, Gordon, Marschner, Steve, Debevec, Paul
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916673430224896
author Deng, Youming
Xian, Wenqi
Yang, Guandao
Guibas, Leonidas
Wetzstein, Gordon
Marschner, Steve
Debevec, Paul
author_facet Deng, Youming
Xian, Wenqi
Yang, Guandao
Guibas, Leonidas
Wetzstein, Gordon
Marschner, Steve
Debevec, Paul
contents In this paper, we present a self-calibrating framework that jointly optimizes camera parameters, lens distortion and 3D Gaussian representations, enabling accurate and efficient scene reconstruction. In particular, our technique enables high-quality scene reconstruction from Large field-of-view (FOV) imagery taken with wide-angle lenses, allowing the scene to be modeled from a smaller number of images. Our approach introduces a novel method for modeling complex lens distortions using a hybrid network that combines invertible residual networks with explicit grids. This design effectively regularizes the optimization process, achieving greater accuracy than conventional camera models. Additionally, we propose a cubemap-based resampling strategy to support large FOV images without sacrificing resolution or introducing distortion artifacts. Our method is compatible with the fast rasterization of Gaussian Splatting, adaptable to a wide variety of camera lens distortion, and demonstrates state-of-the-art performance on both synthetic and real-world datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2502_09563
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Self-Calibrating Gaussian Splatting for Large Field of View Reconstruction
Deng, Youming
Xian, Wenqi
Yang, Guandao
Guibas, Leonidas
Wetzstein, Gordon
Marschner, Steve
Debevec, Paul
Computer Vision and Pattern Recognition
Graphics
In this paper, we present a self-calibrating framework that jointly optimizes camera parameters, lens distortion and 3D Gaussian representations, enabling accurate and efficient scene reconstruction. In particular, our technique enables high-quality scene reconstruction from Large field-of-view (FOV) imagery taken with wide-angle lenses, allowing the scene to be modeled from a smaller number of images. Our approach introduces a novel method for modeling complex lens distortions using a hybrid network that combines invertible residual networks with explicit grids. This design effectively regularizes the optimization process, achieving greater accuracy than conventional camera models. Additionally, we propose a cubemap-based resampling strategy to support large FOV images without sacrificing resolution or introducing distortion artifacts. Our method is compatible with the fast rasterization of Gaussian Splatting, adaptable to a wide variety of camera lens distortion, and demonstrates state-of-the-art performance on both synthetic and real-world datasets.
title Self-Calibrating Gaussian Splatting for Large Field of View Reconstruction
topic Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2502.09563