MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo
Abstract With the advent of 3D Gaussian Splatting (3DGS), new and effective solutions have emerged for 3D reconstruction pipelines and scene representation. However, achieving high-fidelity reconstruction of complex scenes and capturing low-frequency features remain long-standing challenges in the f...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01691-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571178245423104 |
|---|---|
| author | Teng Fei Ligong Bi Jieming Gao Shuixuan Chen Guowei Zhang |
| author_facet | Teng Fei Ligong Bi Jieming Gao Shuixuan Chen Guowei Zhang |
| author_sort | Teng Fei |
| collection | DOAJ |
| description | Abstract With the advent of 3D Gaussian Splatting (3DGS), new and effective solutions have emerged for 3D reconstruction pipelines and scene representation. However, achieving high-fidelity reconstruction of complex scenes and capturing low-frequency features remain long-standing challenges in the field of visual 3D reconstruction. Relying solely on sparse point inputs and simple optimization criteria often leads to non-robust reconstructions of the radiance field, with reconstruction quality heavily dependent on the proper initialization of inputs. Notably, Multi-View Stereo (MVS) techniques offer a mature and reliable approach for generating structured point cloud data using a limited number of views, camera parameters, and feature matching. In this paper, we propose combining MVS with Gaussian Splatting, along with our newly introduced density optimization strategy, to address these challenges. This approach bridges the gap in scene representation by enhancing explicit geometry radiance fields with MVS, and our experimental results demonstrate its effectiveness. Additionally, we have explored the potential of using Gaussian Splatting for non-face template single-process end-to-end Avatar Reconstruction, yielding promising experimental results. |
| format | Article |
| id | doaj-art-82fb2de8a14e4f66a658b558c71d010a |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-82fb2de8a14e4f66a658b558c71d010a2025-02-02T12:49:06ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-12-0111111710.1007/s40747-024-01691-xMVSGS: Gaussian splatting radiation field enhancement using multi-view stereoTeng Fei0Ligong Bi1Jieming Gao2Shuixuan Chen3Guowei Zhang4School of Mechanical and Automotive Engineering, Xiamen University of TechnologyGuilin University of Electronic TechnologyTianjin UniversitySchool of Mechanical and Automotive Engineering, Xiamen University of TechnologySchool of Mechanical and Automotive Engineering, Xiamen University of TechnologyAbstract With the advent of 3D Gaussian Splatting (3DGS), new and effective solutions have emerged for 3D reconstruction pipelines and scene representation. However, achieving high-fidelity reconstruction of complex scenes and capturing low-frequency features remain long-standing challenges in the field of visual 3D reconstruction. Relying solely on sparse point inputs and simple optimization criteria often leads to non-robust reconstructions of the radiance field, with reconstruction quality heavily dependent on the proper initialization of inputs. Notably, Multi-View Stereo (MVS) techniques offer a mature and reliable approach for generating structured point cloud data using a limited number of views, camera parameters, and feature matching. In this paper, we propose combining MVS with Gaussian Splatting, along with our newly introduced density optimization strategy, to address these challenges. This approach bridges the gap in scene representation by enhancing explicit geometry radiance fields with MVS, and our experimental results demonstrate its effectiveness. Additionally, we have explored the potential of using Gaussian Splatting for non-face template single-process end-to-end Avatar Reconstruction, yielding promising experimental results.https://doi.org/10.1007/s40747-024-01691-x3D reconstruction3D Gaussian splattingMulti-view stereoRadiation fieldExplicit geometryDensity optimization |
| spellingShingle | Teng Fei Ligong Bi Jieming Gao Shuixuan Chen Guowei Zhang MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo Complex & Intelligent Systems 3D reconstruction 3D Gaussian splatting Multi-view stereo Radiation field Explicit geometry Density optimization |
| title | MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo |
| title_full | MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo |
| title_fullStr | MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo |
| title_full_unstemmed | MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo |
| title_short | MVSGS: Gaussian splatting radiation field enhancement using multi-view stereo |
| title_sort | mvsgs gaussian splatting radiation field enhancement using multi view stereo |
| topic | 3D reconstruction 3D Gaussian splatting Multi-view stereo Radiation field Explicit geometry Density optimization |
| url | https://doi.org/10.1007/s40747-024-01691-x |
| work_keys_str_mv | AT tengfei mvsgsgaussiansplattingradiationfieldenhancementusingmultiviewstereo AT ligongbi mvsgsgaussiansplattingradiationfieldenhancementusingmultiviewstereo AT jieminggao mvsgsgaussiansplattingradiationfieldenhancementusingmultiviewstereo AT shuixuanchen mvsgsgaussiansplattingradiationfieldenhancementusingmultiviewstereo AT guoweizhang mvsgsgaussiansplattingradiationfieldenhancementusingmultiviewstereo |