DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes
This paper presents an accurate and robust dense 3D reconstruction system for detail preserving surface modeling of large-scale scenes from multi-view images, which we named DP-MVS. Our system performs high-quality large-scale dense reconstruction, which preserves geometric details for thin structur...
Guardado en:
Autores principales: | , , , , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
MDPI AG
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/12aa9427d272436faf18d5ca457a7c29 |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:12aa9427d272436faf18d5ca457a7c29 |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:12aa9427d272436faf18d5ca457a7c292021-11-25T18:54:27ZDP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes10.3390/rs132245692072-4292https://doaj.org/article/12aa9427d272436faf18d5ca457a7c292021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/22/4569https://doaj.org/toc/2072-4292This paper presents an accurate and robust dense 3D reconstruction system for detail preserving surface modeling of large-scale scenes from multi-view images, which we named DP-MVS. Our system performs high-quality large-scale dense reconstruction, which preserves geometric details for thin structures, especially for linear objects. Our framework begins with a sparse reconstruction carried out by an incremental Structure-from-Motion. Based on the reconstructed sparse map, a novel detail preserving PatchMatch approach is applied for depth estimation of each image view. The estimated depth maps of multiple views are then fused to a dense point cloud in a memory-efficient way, followed by a detail-aware surface meshing method to extract the final surface mesh of the captured scene. Experiments on ETH3D benchmark show that the proposed method outperforms other state-of-the-art methods on F1-score, with the running time more than 4 times faster. More experiments on large-scale photo collections demonstrate the effectiveness of the proposed framework for large-scale scene reconstruction in terms of accuracy, completeness, memory saving, and time efficiency.Liyang ZhouZhuang ZhangHanqing JiangHan SunHujun BaoGuofeng ZhangMDPI AGarticlemulti-view reconstructiondetail preservingdepth estimationsurface meshingScienceQENRemote Sensing, Vol 13, Iss 4569, p 4569 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
multi-view reconstruction detail preserving depth estimation surface meshing Science Q |
spellingShingle |
multi-view reconstruction detail preserving depth estimation surface meshing Science Q Liyang Zhou Zhuang Zhang Hanqing Jiang Han Sun Hujun Bao Guofeng Zhang DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
description |
This paper presents an accurate and robust dense 3D reconstruction system for detail preserving surface modeling of large-scale scenes from multi-view images, which we named DP-MVS. Our system performs high-quality large-scale dense reconstruction, which preserves geometric details for thin structures, especially for linear objects. Our framework begins with a sparse reconstruction carried out by an incremental Structure-from-Motion. Based on the reconstructed sparse map, a novel detail preserving PatchMatch approach is applied for depth estimation of each image view. The estimated depth maps of multiple views are then fused to a dense point cloud in a memory-efficient way, followed by a detail-aware surface meshing method to extract the final surface mesh of the captured scene. Experiments on ETH3D benchmark show that the proposed method outperforms other state-of-the-art methods on F1-score, with the running time more than 4 times faster. More experiments on large-scale photo collections demonstrate the effectiveness of the proposed framework for large-scale scene reconstruction in terms of accuracy, completeness, memory saving, and time efficiency. |
format |
article |
author |
Liyang Zhou Zhuang Zhang Hanqing Jiang Han Sun Hujun Bao Guofeng Zhang |
author_facet |
Liyang Zhou Zhuang Zhang Hanqing Jiang Han Sun Hujun Bao Guofeng Zhang |
author_sort |
Liyang Zhou |
title |
DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
title_short |
DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
title_full |
DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
title_fullStr |
DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
title_full_unstemmed |
DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes |
title_sort |
dp-mvs: detail preserving multi-view surface reconstruction of large-scale scenes |
publisher |
MDPI AG |
publishDate |
2021 |
url |
https://doaj.org/article/12aa9427d272436faf18d5ca457a7c29 |
work_keys_str_mv |
AT liyangzhou dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes AT zhuangzhang dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes AT hanqingjiang dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes AT hansun dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes AT hujunbao dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes AT guofengzhang dpmvsdetailpreservingmultiviewsurfacereconstructionoflargescalescenes |
_version_ |
1718410571387240448 |