English

Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference

Computer Vision and Pattern Recognition 2019-02-28 v1

Abstract

Deep learning has recently demonstrated its excellent performance for multi-view stereo (MVS). However, one major limitation of current learned MVS approaches is the scalability: the memory-consuming cost volume regularization makes the learned MVS hard to be applied to high-resolution scenes. In this paper, we introduce a scalable multi-view stereo framework based on the recurrent neural network. Instead of regularizing the entire 3D cost volume in one go, the proposed Recurrent Multi-view Stereo Network (R-MVSNet) sequentially regularizes the 2D cost maps along the depth direction via the gated recurrent unit (GRU). This reduces dramatically the memory consumption and makes high-resolution reconstruction feasible. We first show the state-of-the-art performance achieved by the proposed R-MVSNet on the recent MVS benchmarks. Then, we further demonstrate the scalability of the proposed method on several large-scale scenarios, where previous learned approaches often fail due to the memory constraint. Code is available at https://github.com/YoYo000/MVSNet.

Keywords

Cite

@article{arxiv.1902.10556,
  title  = {Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference},
  author = {Yao Yao and Zixin Luo and Shiwei Li and Tianwei Shen and Tian Fang and Long Quan},
  journal= {arXiv preprint arXiv:1902.10556},
  year   = {2019}
}

Comments

Accepted by CVPR2019

R2 v1 2026-06-23T07:53:03.583Z