L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery Article Swipe

PDF

Yi-Zhen Tsai , Xiaolei Zhang , Jiasi Chen ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2504.05517

Traditional 3D content representations include dense point clouds that consume large amounts of data and hence network bandwidth, while newer representations such as neural radiance fields suffer from poor frame rates due to their non-standard volumetric rendering pipeline. 3D Gaussian splats (3DGS) can be seen as a generalization of point clouds that meet the best of both worlds, with high visual quality and efficient rendering for real-time frame rates. However, delivering 3DGS scenes from a hosting server to client devices is still challenging due to high network data consumption (e.g., 1.5 GB for a single scene). The goal of this work is to create an efficient 3D content delivery framework that allows users to view high quality 3D scenes with 3DGS as the underlying data representation. The main contributions of the paper are: (1) Creating new layered 3DGS scenes for efficient delivery, (2) Scheduling algorithms to choose what splats to download at what time, and (3) Trace-driven experiments from users wearing virtual reality headsets to evaluate the visual quality and latency. Our system for Layered 3D Gaussian Splats delivery L3GS demonstrates high visual quality, achieving 16.9% higher average SSIM compared to baselines, and also works with other compressed 3DGS representations.

Related Topics

Nuclear Weapons Delivery

Crime Scene

Kiki's Delivery Service

Roundhay Garden Scene

Delivery Hero

Selena Gomez & The Scene

Efficient-Market Hypothesis

Palm Desert Scene

Seal Delivery Vehicle

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2504.05517
PDF: https://arxiv.org/pdf/2504.05517
OA Status: green
OpenAlex ID: https://openalex.org/W4416527481

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4416527481

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2504.05517

Digital Object Identifier
Title: L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-04-07

Full publication date if available
Authors: Yi-Zhen Tsai, Xiaolei Zhang, Jiasi Chen

List of authors in order
Landing page: https://arxiv.org/abs/2504.05517

Publisher landing page
PDF URL: https://arxiv.org/pdf/2504.05517

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2504.05517

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4416527481
doi	https://doi.org/10.48550/arxiv.2504.05517
ids.doi	https://doi.org/10.48550/arxiv.2504.05517
ids.openalex	https://openalex.org/W4416527481
fwci
type	preprint
title	L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2504.05517
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2504.05517
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2504.05517
locations[1].id	doi:10.48550/arxiv.2504.05517
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2504.05517
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5061871415
authorships[0].author.orcid
authorships[0].author.display_name	Yi-Zhen Tsai
authorships[0].author_position	first
authorships[0].raw_author_name	Tsai, Yi-Zhen
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5100450095
authorships[1].author.orcid	https://orcid.org/0000-0001-9415-3136
authorships[1].author.display_name	Xiaolei Zhang
authorships[1].author_position	middle
authorships[1].raw_author_name	Zhang, Xuechen
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5061832197
authorships[2].author.orcid	https://orcid.org/0000-0001-9923-9027
authorships[2].author.display_name	Jiasi Chen
authorships[2].author_position	middle
authorships[2].raw_author_name	Chen, Jiasi
authorships[2].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2504.05517
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	L3GS: Layered 3D Gaussian Splats for Efficient 3D Scene Delivery
has_fulltext	False
is_retracted	False
updated_date	2025-11-28T16:34:24.145480
primary_topic
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2504.05517
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2504.05517
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2504.05517
primary_location.id	pmh:oai:arXiv.org:2504.05517
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2504.05517
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2504.05517
publication_date	2025-04-07
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	46, 74, 93
abstract_inverted_index.3D	1, 38, 106, 117, 175
abstract_inverted_index.GB	91
abstract_inverted_index.an	104
abstract_inverted_index.as	22, 45, 121
abstract_inverted_index.at	151
abstract_inverted_index.be	43
abstract_inverted_index.is	80, 101
abstract_inverted_index.of	12, 48, 55, 98, 129
abstract_inverted_index.to	32, 77, 84, 102, 113, 145, 149, 164, 190
abstract_inverted_index.(1)	133
abstract_inverted_index.(2)	142
abstract_inverted_index.(3)	155
abstract_inverted_index.1.5	90
abstract_inverted_index.Our	171
abstract_inverted_index.The	96, 126
abstract_inverted_index.and	14, 62, 154, 169, 192
abstract_inverted_index.can	42
abstract_inverted_index.due	31, 83
abstract_inverted_index.for	65, 92, 139, 173
abstract_inverted_index.new	135
abstract_inverted_index.the	53, 122, 130, 166
abstract_inverted_index.3DGS	71, 120, 137, 198
abstract_inverted_index.L3GS	179
abstract_inverted_index.SSIM	188
abstract_inverted_index.also	193
abstract_inverted_index.are:	132
abstract_inverted_index.best	54
abstract_inverted_index.both	56
abstract_inverted_index.data	13, 87, 124
abstract_inverted_index.from	27, 73, 158
abstract_inverted_index.goal	97
abstract_inverted_index.high	59, 85, 115, 181
abstract_inverted_index.main	127
abstract_inverted_index.meet	52
abstract_inverted_index.poor	28
abstract_inverted_index.seen	44
abstract_inverted_index.such	21
abstract_inverted_index.that	8, 51, 110
abstract_inverted_index.this	99
abstract_inverted_index.view	114
abstract_inverted_index.what	147, 152
abstract_inverted_index.with	58, 119, 195
abstract_inverted_index.work	100
abstract_inverted_index.16.9%	185
abstract_inverted_index.dense	5
abstract_inverted_index.frame	29, 67
abstract_inverted_index.hence	15
abstract_inverted_index.large	10
abstract_inverted_index.newer	19
abstract_inverted_index.other	196
abstract_inverted_index.paper	131
abstract_inverted_index.point	6, 49
abstract_inverted_index.rates	30
abstract_inverted_index.still	81
abstract_inverted_index.their	33
abstract_inverted_index.time,	153
abstract_inverted_index.users	112, 159
abstract_inverted_index.while	18
abstract_inverted_index.works	194
abstract_inverted_index.(3DGS)	41
abstract_inverted_index.(e.g.,	89
abstract_inverted_index.Splats	177
abstract_inverted_index.allows	111
abstract_inverted_index.choose	146
abstract_inverted_index.client	78
abstract_inverted_index.clouds	7, 50
abstract_inverted_index.create	103
abstract_inverted_index.fields	25
abstract_inverted_index.higher	186
abstract_inverted_index.neural	23
abstract_inverted_index.rates.	68
abstract_inverted_index.scenes	72, 118, 138
abstract_inverted_index.server	76
abstract_inverted_index.single	94
abstract_inverted_index.splats	40, 148
abstract_inverted_index.suffer	26
abstract_inverted_index.system	172
abstract_inverted_index.visual	60, 167, 182
abstract_inverted_index.Layered	174
abstract_inverted_index.amounts	11
abstract_inverted_index.average	187
abstract_inverted_index.consume	9
abstract_inverted_index.content	2, 107
abstract_inverted_index.devices	79
abstract_inverted_index.hosting	75
abstract_inverted_index.include	4
abstract_inverted_index.layered	136
abstract_inverted_index.network	16, 86
abstract_inverted_index.quality	61, 116, 168
abstract_inverted_index.reality	162
abstract_inverted_index.scene).	95
abstract_inverted_index.virtual	161
abstract_inverted_index.wearing	160
abstract_inverted_index.worlds,	57
abstract_inverted_index.Creating	134
abstract_inverted_index.Gaussian	39, 176
abstract_inverted_index.However,	69
abstract_inverted_index.compared	189
abstract_inverted_index.delivery	108, 178
abstract_inverted_index.download	150
abstract_inverted_index.evaluate	165
abstract_inverted_index.headsets	163
abstract_inverted_index.latency.	170
abstract_inverted_index.quality,	183
abstract_inverted_index.radiance	24
abstract_inverted_index.achieving	184
abstract_inverted_index.delivery,	141
abstract_inverted_index.efficient	63, 105, 140
abstract_inverted_index.framework	109
abstract_inverted_index.pipeline.	37
abstract_inverted_index.real-time	66
abstract_inverted_index.rendering	36, 64
abstract_inverted_index.Scheduling	143
abstract_inverted_index.algorithms	144
abstract_inverted_index.bandwidth,	17
abstract_inverted_index.baselines,	191
abstract_inverted_index.compressed	197
abstract_inverted_index.delivering	70
abstract_inverted_index.underlying	123
abstract_inverted_index.volumetric	35
abstract_inverted_index.Traditional	0
abstract_inverted_index.challenging	82
abstract_inverted_index.consumption	88
abstract_inverted_index.experiments	157
abstract_inverted_index.Trace-driven	156
abstract_inverted_index.demonstrates	180
abstract_inverted_index.non-standard	34
abstract_inverted_index.contributions	128
abstract_inverted_index.generalization	47
abstract_inverted_index.representation.	125
abstract_inverted_index.representations	3, 20
abstract_inverted_index.representations.	199
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	3
citation_normalized_percentile