IBGS: Image-Based Gaussian Splatting

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Nguyen, Hoang Chuong, Mao, Wei, Alvarez, Jose M., Liu, Miaomiao
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909910912991232
author Nguyen, Hoang Chuong
Mao, Wei
Alvarez, Jose M.
Liu, Miaomiao
author_facet Nguyen, Hoang Chuong
Mao, Wei
Alvarez, Jose M.
Liu, Miaomiao
contents 3D Gaussian Splatting (3DGS) has recently emerged as a fast, high-quality method for novel view synthesis (NVS). However, its use of low-degree spherical harmonics limits its ability to capture spatially varying color and view-dependent effects such as specular highlights. Existing works augment Gaussians with either a global texture map, which struggles with complex scenes, or per-Gaussian texture maps, which introduces high storage overhead. We propose Image-Based Gaussian Splatting, an efficient alternative that leverages high-resolution source images for fine details and view-specific color modeling. Specifically, we model each pixel color as a combination of a base color from standard 3DGS rendering and a learned residual inferred from neighboring training images. This promotes accurate surface alignment and enables rendering images of high-frequency details and accurate view-dependent effects. Experiments on standard NVS benchmarks show that our method significantly outperforms prior Gaussian Splatting approaches in rendering quality, without increasing the storage footprint.
format Preprint
id arxiv_https___arxiv_org_abs_2511_14357
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle IBGS: Image-Based Gaussian Splatting
Nguyen, Hoang Chuong
Mao, Wei
Alvarez, Jose M.
Liu, Miaomiao
Computer Vision and Pattern Recognition
3D Gaussian Splatting (3DGS) has recently emerged as a fast, high-quality method for novel view synthesis (NVS). However, its use of low-degree spherical harmonics limits its ability to capture spatially varying color and view-dependent effects such as specular highlights. Existing works augment Gaussians with either a global texture map, which struggles with complex scenes, or per-Gaussian texture maps, which introduces high storage overhead. We propose Image-Based Gaussian Splatting, an efficient alternative that leverages high-resolution source images for fine details and view-specific color modeling. Specifically, we model each pixel color as a combination of a base color from standard 3DGS rendering and a learned residual inferred from neighboring training images. This promotes accurate surface alignment and enables rendering images of high-frequency details and accurate view-dependent effects. Experiments on standard NVS benchmarks show that our method significantly outperforms prior Gaussian Splatting approaches in rendering quality, without increasing the storage footprint.
title IBGS: Image-Based Gaussian Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.14357