3DGEER: 3D Gaussian Rendering Made Exact and Efficient for Generic Cameras

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Huang, Zixun, Wu, Cho-Ying, Guo, Yuliang, Huang, Xinyu, Ren, Liu
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918383913533440
author Huang, Zixun
Wu, Cho-Ying
Guo, Yuliang
Huang, Xinyu
Ren, Liu
author_facet Huang, Zixun
Wu, Cho-Ying
Guo, Yuliang
Huang, Xinyu
Ren, Liu
contents 3D Gaussian Splatting (3DGS) achieves an appealing balance between rendering quality and efficiency, but relies on approximating 3D Gaussians as 2D projections--an assumption that degrades accuracy, especially under generic large field-of-view (FoV) cameras. Despite recent extensions, no prior work has simultaneously achieved both projective exactness and real-time efficiency for general cameras. We introduce 3DGEER, a geometrically exact and efficient Gaussian rendering framework. From first principles, we derive a closed-form expression for integrating Gaussian density along a ray, enabling precise forward rendering and differentiable optimization under arbitrary camera models. To retain efficiency, we propose the Particle Bounding Frustum (PBF), which provides tight ray-Gaussian association without BVH traversal, and the Bipolar Equiangular Projection (BEAP), which unifies FoV representations, accelerates association, and improves reconstruction quality. Experiments on both pinhole and fisheye datasets show that 3DGEER outperforms prior methods across all metrics, runs 5x faster than existing projective exact ray-based baselines, and generalizes to wider FoVs unseen during training--establishing a new state of the art in real-time radiance field rendering.
format Preprint
id arxiv_https___arxiv_org_abs_2505_24053
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle 3DGEER: 3D Gaussian Rendering Made Exact and Efficient for Generic Cameras
Huang, Zixun
Wu, Cho-Ying
Guo, Yuliang
Huang, Xinyu
Ren, Liu
Graphics
Computer Vision and Pattern Recognition
I.3.7; I.2.10
3D Gaussian Splatting (3DGS) achieves an appealing balance between rendering quality and efficiency, but relies on approximating 3D Gaussians as 2D projections--an assumption that degrades accuracy, especially under generic large field-of-view (FoV) cameras. Despite recent extensions, no prior work has simultaneously achieved both projective exactness and real-time efficiency for general cameras. We introduce 3DGEER, a geometrically exact and efficient Gaussian rendering framework. From first principles, we derive a closed-form expression for integrating Gaussian density along a ray, enabling precise forward rendering and differentiable optimization under arbitrary camera models. To retain efficiency, we propose the Particle Bounding Frustum (PBF), which provides tight ray-Gaussian association without BVH traversal, and the Bipolar Equiangular Projection (BEAP), which unifies FoV representations, accelerates association, and improves reconstruction quality. Experiments on both pinhole and fisheye datasets show that 3DGEER outperforms prior methods across all metrics, runs 5x faster than existing projective exact ray-based baselines, and generalizes to wider FoVs unseen during training--establishing a new state of the art in real-time radiance field rendering.
title 3DGEER: 3D Gaussian Rendering Made Exact and Efficient for Generic Cameras
topic Graphics
Computer Vision and Pattern Recognition
I.3.7; I.2.10
url https://arxiv.org/abs/2505.24053