Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss Article Swipe
Wen Feng Lu
,
Haodong Chen
,
Anqi Yi
,
Yuk Ying Chung
,
Zhiyong Wang
,
Kun Hu
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2505.22279
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2505.22279
Novel view synthesis is a fundamental task in 3D computer vision that aims to reconstruct realistic images from a set of posed input views. However, reconstruction quality degrades significantly under sparse-view conditions due to limited geometric cues. Existing methods, such as Neural Radiance Fields (NeRF) and the more recent 3D Gaussian Splatting (3DGS), often suffer from blurred details and structural artifacts when trained with insufficient views. Recent works have identified the quality of rendered depth as a key factor in mitigating these artifacts, as it directly affects geometric accuracy and view consistency. In this paper, we address these challenges by introducing Hierarchical Depth-Guided Splatting (HDGS), a depth supervision framework that progressively refines geometry from coarse to fine levels. Central to HDGS is a novel Cascade Pearson Correlation Loss (CPCL), which aligns rendered and estimated monocular depths across multiple spatial scales. By enforcing multi-scale depth consistency, our method substantially improves structural fidelity in sparse-view scenarios. Extensive experiments on the LLFF and DTU benchmarks demonstrate that HDGS achieves state-of-the-art performance under sparse-view settings while maintaining efficient and high-quality rendering
Related Topics
Concepts
No concepts available.
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2505.22279
- https://arxiv.org/pdf/2505.22279
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4416048011
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4416048011Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2505.22279Digital Object Identifier
- Title
-
Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth LossWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-05-28Full publication date if available
- Authors
-
Wen Feng Lu, Haodong Chen, Anqi Yi, Yuk Ying Chung, Zhiyong Wang, Kun HuList of authors in order
- Landing page
-
https://arxiv.org/abs/2505.22279Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2505.22279Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2505.22279Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4416048011 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2505.22279 |
| ids.doi | https://doi.org/10.48550/arxiv.2505.22279 |
| ids.openalex | https://openalex.org/W4416048011 |
| fwci | |
| type | preprint |
| title | Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss |
| 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:2505.22279 |
| 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 | cc-by-nc-sa |
| locations[0].pdf_url | https://arxiv.org/pdf/2505.22279 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/cc-by-nc-sa |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2505.22279 |
| locations[1].id | doi:10.48550/arxiv.2505.22279 |
| 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 | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| 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.2505.22279 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5010686710 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-4022-6912 |
| authorships[0].author.display_name | Wen Feng Lu |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Lu, Wenjun |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5101495668 |
| authorships[1].author.orcid | https://orcid.org/0000-0003-2254-5629 |
| authorships[1].author.display_name | Haodong Chen |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Chen, Haodong |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5031201714 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Anqi Yi |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Yi, Anqi |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5045843249 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-3158-9650 |
| authorships[3].author.display_name | Yuk Ying Chung |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Chung, Yuk Ying |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5100443939 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-0099-7587 |
| authorships[4].author.display_name | Zhiyong Wang |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Wang, Zhiyong |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5006591448 |
| authorships[5].author.orcid | https://orcid.org/0009-0002-0831-8059 |
| authorships[5].author.display_name | Kun Hu |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Hu, Kun |
| authorships[5].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2505.22279 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-28T10:59:56.676101 |
| primary_topic | |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2505.22279 |
| 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 | cc-by-nc-sa |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2505.22279 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by-nc-sa |
| 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/2505.22279 |
| primary_location.id | pmh:oai:arXiv.org:2505.22279 |
| 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 | cc-by-nc-sa |
| primary_location.pdf_url | https://arxiv.org/pdf/2505.22279 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/cc-by-nc-sa |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2505.22279 |
| publication_date | 2025-05-28 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 4, 18, 76, 105, 122 |
| abstract_inverted_index.3D | 8, 49 |
| abstract_inverted_index.By | 140 |
| abstract_inverted_index.In | 92 |
| abstract_inverted_index.as | 40, 75, 83 |
| abstract_inverted_index.by | 99 |
| abstract_inverted_index.in | 7, 79, 151 |
| abstract_inverted_index.is | 3, 121 |
| abstract_inverted_index.it | 84 |
| abstract_inverted_index.of | 20, 72 |
| abstract_inverted_index.on | 156 |
| abstract_inverted_index.to | 13, 33, 115, 119 |
| abstract_inverted_index.we | 95 |
| abstract_inverted_index.DTU | 160 |
| abstract_inverted_index.and | 45, 58, 89, 132, 159, 174 |
| abstract_inverted_index.due | 32 |
| abstract_inverted_index.key | 77 |
| abstract_inverted_index.our | 145 |
| abstract_inverted_index.set | 19 |
| abstract_inverted_index.the | 46, 70, 157 |
| abstract_inverted_index.HDGS | 120, 164 |
| abstract_inverted_index.LLFF | 158 |
| abstract_inverted_index.Loss | 127 |
| abstract_inverted_index.aims | 12 |
| abstract_inverted_index.fine | 116 |
| abstract_inverted_index.from | 17, 55, 113 |
| abstract_inverted_index.have | 68 |
| abstract_inverted_index.more | 47 |
| abstract_inverted_index.such | 39 |
| abstract_inverted_index.task | 6 |
| abstract_inverted_index.that | 11, 109, 163 |
| abstract_inverted_index.this | 93 |
| abstract_inverted_index.view | 1, 90 |
| abstract_inverted_index.when | 61 |
| abstract_inverted_index.with | 63 |
| abstract_inverted_index.Novel | 0 |
| abstract_inverted_index.cues. | 36 |
| abstract_inverted_index.depth | 74, 106, 143 |
| abstract_inverted_index.input | 22 |
| abstract_inverted_index.novel | 123 |
| abstract_inverted_index.often | 53 |
| abstract_inverted_index.posed | 21 |
| abstract_inverted_index.these | 81, 97 |
| abstract_inverted_index.under | 29, 168 |
| abstract_inverted_index.which | 129 |
| abstract_inverted_index.while | 171 |
| abstract_inverted_index.works | 67 |
| abstract_inverted_index.(NeRF) | 44 |
| abstract_inverted_index.Fields | 43 |
| abstract_inverted_index.Neural | 41 |
| abstract_inverted_index.Recent | 66 |
| abstract_inverted_index.across | 136 |
| abstract_inverted_index.aligns | 130 |
| abstract_inverted_index.coarse | 114 |
| abstract_inverted_index.depths | 135 |
| abstract_inverted_index.factor | 78 |
| abstract_inverted_index.images | 16 |
| abstract_inverted_index.method | 146 |
| abstract_inverted_index.paper, | 94 |
| abstract_inverted_index.recent | 48 |
| abstract_inverted_index.suffer | 54 |
| abstract_inverted_index.views. | 23, 65 |
| abstract_inverted_index.vision | 10 |
| abstract_inverted_index.(3DGS), | 52 |
| abstract_inverted_index.(CPCL), | 128 |
| abstract_inverted_index.(HDGS), | 104 |
| abstract_inverted_index.Cascade | 124 |
| abstract_inverted_index.Central | 118 |
| abstract_inverted_index.Pearson | 125 |
| abstract_inverted_index.address | 96 |
| abstract_inverted_index.affects | 86 |
| abstract_inverted_index.blurred | 56 |
| abstract_inverted_index.details | 57 |
| abstract_inverted_index.levels. | 117 |
| abstract_inverted_index.limited | 34 |
| abstract_inverted_index.quality | 26, 71 |
| abstract_inverted_index.refines | 111 |
| abstract_inverted_index.scales. | 139 |
| abstract_inverted_index.spatial | 138 |
| abstract_inverted_index.trained | 62 |
| abstract_inverted_index.Existing | 37 |
| abstract_inverted_index.Gaussian | 50 |
| abstract_inverted_index.However, | 24 |
| abstract_inverted_index.Radiance | 42 |
| abstract_inverted_index.accuracy | 88 |
| abstract_inverted_index.achieves | 165 |
| abstract_inverted_index.computer | 9 |
| abstract_inverted_index.degrades | 27 |
| abstract_inverted_index.directly | 85 |
| abstract_inverted_index.fidelity | 150 |
| abstract_inverted_index.geometry | 112 |
| abstract_inverted_index.improves | 148 |
| abstract_inverted_index.methods, | 38 |
| abstract_inverted_index.multiple | 137 |
| abstract_inverted_index.rendered | 73, 131 |
| abstract_inverted_index.settings | 170 |
| abstract_inverted_index.Extensive | 154 |
| abstract_inverted_index.Splatting | 51, 103 |
| abstract_inverted_index.artifacts | 60 |
| abstract_inverted_index.efficient | 173 |
| abstract_inverted_index.enforcing | 141 |
| abstract_inverted_index.estimated | 133 |
| abstract_inverted_index.framework | 108 |
| abstract_inverted_index.geometric | 35, 87 |
| abstract_inverted_index.monocular | 134 |
| abstract_inverted_index.realistic | 15 |
| abstract_inverted_index.rendering | 176 |
| abstract_inverted_index.synthesis | 2 |
| abstract_inverted_index.artifacts, | 82 |
| abstract_inverted_index.benchmarks | 161 |
| abstract_inverted_index.challenges | 98 |
| abstract_inverted_index.conditions | 31 |
| abstract_inverted_index.identified | 69 |
| abstract_inverted_index.mitigating | 80 |
| abstract_inverted_index.scenarios. | 153 |
| abstract_inverted_index.structural | 59, 149 |
| abstract_inverted_index.Correlation | 126 |
| abstract_inverted_index.demonstrate | 162 |
| abstract_inverted_index.experiments | 155 |
| abstract_inverted_index.fundamental | 5 |
| abstract_inverted_index.introducing | 100 |
| abstract_inverted_index.maintaining | 172 |
| abstract_inverted_index.multi-scale | 142 |
| abstract_inverted_index.performance | 167 |
| abstract_inverted_index.reconstruct | 14 |
| abstract_inverted_index.sparse-view | 30, 152, 169 |
| abstract_inverted_index.supervision | 107 |
| abstract_inverted_index.Depth-Guided | 102 |
| abstract_inverted_index.Hierarchical | 101 |
| abstract_inverted_index.consistency, | 144 |
| abstract_inverted_index.consistency. | 91 |
| abstract_inverted_index.high-quality | 175 |
| abstract_inverted_index.insufficient | 64 |
| abstract_inverted_index.progressively | 110 |
| abstract_inverted_index.significantly | 28 |
| abstract_inverted_index.substantially | 147 |
| abstract_inverted_index.reconstruction | 25 |
| abstract_inverted_index.state-of-the-art | 166 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 6 |
| citation_normalized_percentile |