DeepMVS is a deep learning based 3D Reconstruction project that is based on Colmap for SfM, MVSNet for depth infering, and OpenMVS for depth refine, depth to pointcloud, pointcloud to mesh.
We are much faser than OpenMVS with the learning based depth infering network.
- OS: Ubuntu 16.04 or 18.04
- NVIDIA GPU with CUDA>=10.0
For OpenMVS: Please refer to OpenMVS
- Eigen version 3.2 (or higher on Windows only)
- OpenCV version 2.4 or higher
- Ceres version 1.10 or higher
- CGAL version 4.2 or higher
- Boost version 1.56 or higher
- VCG
- GLFW
For CasMVSNet_pl: Please refer to CasMVSNet_pl which is a variant of MVSNet
- Python>=3.6.1
- pytorch>=1.4
- pytorch_lightning
Reconstruction:
Run (example)
bash test.sh
TODO: add the complete codes