ACMM

ACMM is a multi-scale geometric consistency guided multi-view stereo method for efficient and accurate depth map estimation. If you find this project useful for your research, please cite:

@article{Xu2019ACMM,  
  title={Multi-Scale Geometric Consistency Guided Multi-View Stereo}, 
  author={Xu, Qingshan and Tao, Wenbing}, 
  journal={Computer Vision and Pattern Recognition (CVPR)},
  year={2019}
}

Dependencies

The code has been tested on Ubuntu 14.04 with GTX Titan X.

Usage

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

Results on high-res ETH3D training dataset [2cm]

Mean courtyard delivery_area electro facade kicker meadow office pipes playgroud relief relief_2 terrace terrains
78.87 87.47 83.70 86.82 70.79 77.51 66.41 64.88 70.07 72.06 85.36 84.87 89.85 85.52

SfM Reconstructions for Tanks and Temples Dataset

To ease comparison with other MVS methods with our method on Tanks and Temples dataset, we release our SfM reconstuctions on this dataset. They are obtained by COLMAP and can be downloaded from here.

Acknowledgements

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

GitHub

https://github.com/GhiXu/ACMM