View in EDS

CoSurfGS: 3D Surface Gaussian Splatting with Collaborative Distributed Learning for Large-scale Scene Reconstruction.

Saved in:
Bibliographic Details
Title: CoSurfGS: 3D Surface Gaussian Splatting with Collaborative Distributed Learning for Large-scale Scene Reconstruction.
Authors: Gao, Yuanyuan1 (AUTHOR) yygao7645@gmail.com, Dai, Yalun2 (AUTHOR) dialogue_dylan@outlook.com, Li, Hao1 (AUTHOR) lifugan_10027@outlook.com, Ye, Weicai3 (AUTHOR) maikeyeweicai@gmail.com, Chen, Junyi4 (AUTHOR), Chen, Danpeng3 (AUTHOR), Zhang, Dingwen1 (AUTHOR) zhangdingwen2006yyy@gmail.com, He, Tong4 (AUTHOR), Zhang, Guofeng3 (AUTHOR), Han, Junwei5 (AUTHOR)
Source: International Journal of Computer Vision. May2026, Vol. 134 Issue 5, p1-14. 14p.
Subjects: Three-dimensional modeling, Rendering (Computer graphics), Ensemble learning, Federated learning, Three-dimensional imaging
Abstract: 3D Gaussian Splatting (3DGS) has demonstrated impressive performance in scene reconstruction. However, most existing GS-based surface reconstruction methods focus on 3D objects or scenes with limited scale. Directly applying these methods to large-scale scene reconstruction will pose challenges such as high memory costs, excessive time consumption, and lack of geometric detail, which makes it difficult to implement in practical applications. 3D Gaussian Splatting (3DGS) has demonstrated impressive performance in scene reconstruction. However, most existing GS-based surface reconstruction methods are limited to single-machine settings and focus on individual objects or scenes with limited scale. When extended to large-scale scene reconstruction, these methods suffer from high memory consumption, prolonged training time, and insufficient geometric detail, which makes it difficult to implement in practical applications. To overcome these limitations, a natural solution is to distribute the learning workload across multiple collaborative devices. However, collaborative distributed learning presents unique challenges, including efficiently training models on resource-limited local devices and effectively integrating knowledge across devices to maintain global consistency and high-quality reconstruction. To address these challenges, we propose CoSurfGS, a novel collaborative-distributed framework that enables high-quality large-scale surface reconstruction while maintaining efficient training and GPU memory utilization. Specifically, we propose two modules: Local Model Compression (LMC), which eliminates redundant Gaussians to improve memory and training efficiency on each device; and Model Aggregation Schemes (MAS), which enhance global reconstruction quality by collaboratively distilling knowledge from multiple distributed devices. Extensive experiments on Urban3d, MegaNeRF, and BlendedMVS demonstrate that our proposed method can achieve fast and scalable high-fidelity surface reconstruction and photorealistic rendering. [ABSTRACT FROM AUTHOR]
Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 192788054
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: CoSurfGS: 3D Surface Gaussian Splatting with Collaborative Distributed Learning for Large-scale Scene Reconstruction.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Gao%2C+Yuanyuan%22">Gao, Yuanyuan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yygao7645@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Dai%2C+Yalun%22">Dai, Yalun</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> dialogue_dylan@outlook.com</i><br /><searchLink fieldCode="AR" term="%22Li%2C+Hao%22">Li, Hao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> lifugan_10027@outlook.com</i><br /><searchLink fieldCode="AR" term="%22Ye%2C+Weicai%22">Ye, Weicai</searchLink><relatesTo>3</relatesTo> (AUTHOR)<i> maikeyeweicai@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Chen%2C+Junyi%22">Chen, Junyi</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Chen%2C+Danpeng%22">Chen, Danpeng</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Dingwen%22">Zhang, Dingwen</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zhangdingwen2006yyy@gmail.com</i><br /><searchLink fieldCode="AR" term="%22He%2C+Tong%22">He, Tong</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Guofeng%22">Zhang, Guofeng</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Han%2C+Junwei%22">Han, Junwei</searchLink><relatesTo>5</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22International+Journal+of+Computer+Vision%22">International Journal of Computer Vision</searchLink>. May2026, Vol. 134 Issue 5, p1-14. 14p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Three-dimensional+modeling%22">Three-dimensional modeling</searchLink><br /><searchLink fieldCode="DE" term="%22Rendering+%28Computer+graphics%29%22">Rendering (Computer graphics)</searchLink><br /><searchLink fieldCode="DE" term="%22Ensemble+learning%22">Ensemble learning</searchLink><br /><searchLink fieldCode="DE" term="%22Federated+learning%22">Federated learning</searchLink><br /><searchLink fieldCode="DE" term="%22Three-dimensional+imaging%22">Three-dimensional imaging</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: 3D Gaussian Splatting (3DGS) has demonstrated impressive performance in scene reconstruction. However, most existing GS-based surface reconstruction methods focus on 3D objects or scenes with limited scale. Directly applying these methods to large-scale scene reconstruction will pose challenges such as high memory costs, excessive time consumption, and lack of geometric detail, which makes it difficult to implement in practical applications. 3D Gaussian Splatting (3DGS) has demonstrated impressive performance in scene reconstruction. However, most existing GS-based surface reconstruction methods are limited to single-machine settings and focus on individual objects or scenes with limited scale. When extended to large-scale scene reconstruction, these methods suffer from high memory consumption, prolonged training time, and insufficient geometric detail, which makes it difficult to implement in practical applications. To overcome these limitations, a natural solution is to distribute the learning workload across multiple collaborative devices. However, collaborative distributed learning presents unique challenges, including efficiently training models on resource-limited local devices and effectively integrating knowledge across devices to maintain global consistency and high-quality reconstruction. To address these challenges, we propose CoSurfGS, a novel collaborative-distributed framework that enables high-quality large-scale surface reconstruction while maintaining efficient training and GPU memory utilization. Specifically, we propose two modules: Local Model Compression (LMC), which eliminates redundant Gaussians to improve memory and training efficiency on each device; and Model Aggregation Schemes (MAS), which enhance global reconstruction quality by collaboratively distilling knowledge from multiple distributed devices. Extensive experiments on Urban3d, MegaNeRF, and BlendedMVS demonstrate that our proposed method can achieve fast and scalable high-fidelity surface reconstruction and photorealistic rendering. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=192788054
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s11263-025-02627-9
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 1
    Subjects:
      – SubjectFull: Three-dimensional modeling
        Type: general
      – SubjectFull: Rendering (Computer graphics)
        Type: general
      – SubjectFull: Ensemble learning
        Type: general
      – SubjectFull: Federated learning
        Type: general
      – SubjectFull: Three-dimensional imaging
        Type: general
    Titles:
      – TitleFull: CoSurfGS: 3D Surface Gaussian Splatting with Collaborative Distributed Learning for Large-scale Scene Reconstruction.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Gao, Yuanyuan
      – PersonEntity:
          Name:
            NameFull: Dai, Yalun
      – PersonEntity:
          Name:
            NameFull: Li, Hao
      – PersonEntity:
          Name:
            NameFull: Ye, Weicai
      – PersonEntity:
          Name:
            NameFull: Chen, Junyi
      – PersonEntity:
          Name:
            NameFull: Chen, Danpeng
      – PersonEntity:
          Name:
            NameFull: Zhang, Dingwen
      – PersonEntity:
          Name:
            NameFull: He, Tong
      – PersonEntity:
          Name:
            NameFull: Zhang, Guofeng
      – PersonEntity:
          Name:
            NameFull: Han, Junwei
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 09205691
          Numbering:
            – Type: volume
              Value: 134
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: International Journal of Computer Vision
              Type: main
ResultId 1