SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsis

Mengqi Ji, Juergen Gall, Haitian Zheng, Yebin Liu, Lu Fang
8/5/2017
cs.CV

Official code: mjiUST/SurfaceNet (124 stars, last commit 2020-07-14).

Abstract

This paper proposes an end-to-end learning framework for multiview stereopsis. We term the network SurfaceNet. It takes a set of images and their corresponding camera parameters as input and directly infers the 3D model. The key advantage of the framework is that both photo-consistency as well geometric relations of the surface structure can be directly learned for the purpose of multiview stereopsis in an end-to-end fashion. SurfaceNet is a fully 3D convolutional network which is achieved by encoding the camera parameters together with the images in a 3D voxel representation. We evaluate SurfaceNet on the large-scale DTU benchmark.

View on arXivView PDF

Code Implementations(1)

mjiUST/SurfaceNetOfficial100%

2017 ICCV, SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsis

12433Shell, MATLABAug 4, 20176 years agoNOASSERTION

Similar Papers

Neural apparent BRDF fields for multiview photometric stereo

84% match2022

CHOSEN: Contrastive Hypothesis Selection for Multi-View Depth Refinement

83% match2024

nLMVS-Net: Deep Non-Lambertian Multi-View Stereo

83% match1 repo72022

Multi-View Stereo Representation Revisit: Region-Aware MVSNet

83% match2023

Multi-View Stereo Representation Revist: Region-Aware MVSNet

83% match2023

Prism: Semi-Supervised Multi-View Stereo with Monocular Structure Priors

83% match1 repo12024

Cite this paper

@article{ji2017surfacenet,
  title  = {SurfaceNet: An End-to-end 3D Neural Network for Multiview Stereopsis},
  author = {Mengqi Ji and Juergen Gall and Haitian Zheng and Yebin Liu and Lu Fang},
  year   = {2017},
  eprint = {1708.01749},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1708.01749}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1708.01749.svg)](https://codeofpaper.com/paper/1708.01749)

Discussion