Mesh-Guided Multi-View Stereo With Pyramid Architecture

Wang, Yuesong; Guan, Tao; Chen, Zhuo; Luo, Yawei; Luo, Keyang; Ju, Lili

doi:10.1109/cvpr42600.2020.00211

Cited by 28 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wei et al [33] proposed a novel selective joint bilateral upsampling and depth propagation strategy for high-resolution unstructured MVS. Wang et al [34] proposed a mesh-guided MVS method with pyramid architecture, which uses the surface mesh obtained from coarsescale images to guide the reconstruction process. However, these methods do not consider too much about how to preserve the true geometric details in depth map estimation and fusion stages.…”

Section: Depth-map Merging Based Methodsmentioning

confidence: 99%

DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes

et al. 2021

View full text Add to dashboard Cite

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.

show abstract

Section: Depth-map Merging Based Methodsmentioning

confidence: 99%

DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Textureless areas are handled with multi-scale geometric consistency guidance. In a similar fashion, [24] added direct planar priors using a probabilistic graphical model whereas [69] used a pyramid architecture and coarse to fine MVS. However, such multi-scale schemes often fail to preserve details.…”

Section: Prior-assisted Patchmatchmentioning

confidence: 99%

Semantically Derived Geometric Constraints for MVS Reconstruction of Textureless Areas

Stathopoulou

Battisti

Cernea³

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Conventional multi-view stereo (MVS) approaches based on photo-consistency measures are generally robust, yet often fail in calculating valid depth pixel estimates in low textured areas of the scene. In this study, a novel approach is proposed to tackle this challenge by leveraging semantic priors into a PatchMatch-based MVS in order to increase confidence and support depth and normal map estimation. Semantic class labels on image pixels are used to impose class-specific geometric constraints during multiview stereo, optimising the depth estimation on weakly supported, textureless areas, commonly present in urban scenarios of building facades, indoor scenes, or aerial datasets. Detecting dominant shapes, e.g., planes, with RANSAC, an adjusted cost function is introduced that combines and weighs both photometric and semantic scores propagating, thus, more accurate depth estimates. Being adaptive, it fills in apparent information gaps and smoothing local roughness in problematic regions while at the same time preserves important details. Experiments on benchmark and custom datasets demonstrate the effectiveness of the presented approach.

show abstract

“…While MVSNet [41] focuses on the reconstruction of 3D scene, DPSNet [12] focuses on evaluating the per-view depth-map accuracy. Researchers have also explored other 3D representations to regularized the prediction, such as point clouds [3], surface normals [15], or meshes [36]. There are also several benchmark datasets for this task [43,37,28,1,5,33].…”

Section: Related Workmentioning

confidence: 99%

MVS2D: Efficient Multi-view Stereo via Attention-Driven 2D Convolutions

Yang

Ren²,

Qi³

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep learning has made significant impacts on multiview stereo systems. State-of-the-art approaches typically involve building a cost volume, followed by multiple 3D convolution operations to recover the input image's pixelwise depth. While such end-to-end learning of planesweeping stereo advances public benchmarks' accuracy, they are typically very slow to compute. We present MVS2D, a highly efficient multi-view stereo algorithm that seamlessly integrates multi-view constraints into single-view networks via an attention mechanism. Since MVS2D only builds on 2D convolutions, it is at least 4× faster than all the notable counterparts. Moreover, our algorithm produces precise depth estimations, achieving state-of-the-art results on challenging benchmarks ScanNet, SUN3D, and RGBD. Even under inexact camera poses, our algorithm still out-performs all other algorithms. Supplementary materials and code will be available at the project page: https://zhenpeiyang.github.io/MVS2D

show abstract

Mesh-Guided Multi-View Stereo With Pyramid Architecture

Cited by 28 publications

References 30 publications

DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes

DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes

Semantically Derived Geometric Constraints for MVS Reconstruction of Textureless Areas

MVS2D: Efficient Multi-view Stereo via Attention-Driven 2D Convolutions

Contact Info

Product

Resources

About