/ACMMP

Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo (TPAMI 2022)

Primary LanguageC++MIT LicenseMIT

ACMMP

[News] The code for ACMH is released!!!
[News] The code for ACMM is released!!!
[News] The code for ACMP is released!!!

About

This repository contains the code for Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo, which is the extension of ACMM and ACMP. If you find our work useful in your research, please consider citing:

@article{Xu2022Multi,
  title={Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo},
  author={Xu, Qingshan and Kong, Weihang and Tao, Wenbing and Pollefeys, Marc},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2022},
  publisher={IEEE}
}

Dependencies

The code has been tested on Ubuntu 16.04 with GTX 1080 Ti.

Usage

  • Complie ACMMP
cmake .  
make
  • Test
Use script colmap2mvsnet_acm.py to convert COLMAP SfM result to ACMMP input   
Run ./ACMMP $data_folder to get reconstruction results

Acknowledgemets

This code largely benefits from the following repositories: Gipuma and COLMAP. Thanks to their authors for opening source of their excellent works.