[News] BlendedMVS dataset is released!!! (project link).
MVSNet is a deep learning architecture for depth map inference from unstructured multi-view images, and R-MVSNet is its extension for scalable learning-based MVS reconstruction. If you find this project useful for your research, please cite:
@article{yao2018mvsnet,
title={MVSNet: Depth Inference for Unstructured Multi-view Stereo},
author={Yao, Yao and Luo, Zixin and Li, Shiwei and Fang, Tian and Quan, Long},
journal={European Conference on Computer Vision (ECCV)},
year={2018}
}
@article{yao2019recurrent,
title={Recurrent MVSNet for High-resolution Multi-view Stereo Depth Inference},
author={Yao, Yao and Luo, Zixin and Li, Shiwei and Shen, Tianwei and Fang, Tian and Quan, Long},
journal={Computer Vision and Pattern Recognition (CVPR)},
year={2019}
}
If BlendedMVS dataset is used in your research, please also cite:
@article{yao2020blendedmvs,
title={BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networks},
author={Yao, Yao and Luo, Zixin and Li, Shiwei and Zhang, Jingyang and Ren, Yufan and Zhou, Lei and Fang, Tian and Quan, Long},
journal={Computer Vision and Pattern Recognition (CVPR)},
year={2020}
}
- Check out the source code
git clone https://github.com/YoYo000/MVSNet
- Install cuda 9.0, cudnn 7.0 and python 2.7
- Install Tensorflow and other dependencies by
sudo pip install -r requirements.txt
- Preprocessed training/validation data: BlendedMVS, DTU and ETH3D. More training resources could be found in BlendedMVS github page
- Preprocessed testing data: DTU testing set, ETH3D testing set, Tanks and Temples testing set and training set
- Pretrained models: pretrained on BlendedMVS, on DTU and on ETH3D
- Enter mvsnet script folder:
cd MVSNet/mvsnet
- Train MVSNet on BlendedMVS, DTU and ETH3D:
python train.py --regularization '3DCNNs' --train_blendedmvs --max_w 768 --max_h 576 --max_d 128 --online_augmentation
python train.py --regularization '3DCNNs' --train_dtu --max_w 640 --max_h 512 --max_d 128
python train.py --regularization '3DCNNs' --train_eth3d --max_w 896 --max_h 480 --max_d 128
- Train R-MVSNet:
python train.py --regularization 'GRU' --train_blendedmvs --max_w 768 --max_h 576 --max_d 128 --online_augmentation
python train.py --regularization 'GRU' --train_dtu --max_w 640 --max_h 512 --max_d 128
python train.py --regularization 'GRU' --train_eth3d --max_w 896 --max_h 480 --max_d 128
- Specify your input training data folders using
--blendedmvs_data_root
,--dtu_data_root
and--eth3d_data_root
- Specify your output log and model folders using
--log_folder
and--model_folder
- Switch from BlendeMVS to BlendedMVG by replacing using
--train_blendedmvs
with--train_blendedmvg
- Validate MVSNet on BlendedMVS, DTU and ETH3D:
python validate.py --regularization '3DCNNs' --validate_set blendedmvs --max_w 768 --max_h 576 --max_d 128
python validate.py --regularization '3DCNNs' --validate_set dtu --max_w 640 --max_h 512 --max_d 128
python validate.py --regularization '3DCNNs' --validate_set eth3d --max_w 896 --max_h 480 --max_d 128
- Validate R-MVSNet:
python validate.py --regularization 'GRU' --validate_set blendedmvs --max_w 768 --max_h 576 --max_d 128
python validate.py --regularization 'GRU' --validate_set dtu --max_w 640 --max_h 512 --max_d 128
python validate.py --regularization 'GRU' --validate_set eth3d --max_w 896 --max_h 480 --max_d 128
- Specify your input model check point using
--pretrained_model_ckpt_path
and--ckpt_step
- Specify your input training data folders using
--blendedmvs_data_root
,--dtu_data_root
and--eth3d_data_root
- Specify your output result file using
--validation_result_path
- Download test data scan9 and unzip it to
TEST_DATA_FOLDER
folder - Run MVSNet (GTX1080Ti):
python test.py --dense_folder TEST_DATA_FOLDER --regularization '3DCNNs' --max_w 1152 --max_h 864 --max_d 192 --interval_scale 1.06
- Run R-MVSNet (GTX1080Ti):
python test.py --dense_folder TEST_DATA_FOLDER --regularization 'GRU' --max_w 1600 --max_h 1200 --max_d 256 --interval_scale 0.8
- Specify your input model check point using
--pretrained_model_ckpt_path
and--ckpt_step
- Specify your input dense folder using
--dense_folder
- Inspect the .pfm format outputs in
TEST_DATA_FOLDER/depths_mvsnet
usingpython visualize.py .pfm
. For example, the depth map and probability map for image00000012
should look like:
![]() |
![]() |
![]() |
---|---|---|
reference image | depth map | probability map |
R/MVSNet itself only produces per-view depth maps. To generate the 3D point cloud, we need to apply depth map filter/fusion for post-processing. As our implementation of this part is depended on the Altizure internal library, currently we could not provide the corresponding code. Fortunately, depth map filter/fusion is a general step in MVS reconstruction, and there are similar implementations in other open-source MVS algorithms. We provide the script depthfusion.py
to utilize fusibile for post-processing (thank Silvano Galliani for the excellent code!).
To run the post-processing:
- Check out the modified version fusibile
git clone https://github.com/YoYo000/fusibile
- Install fusibile by
cmake .
andmake
, which will generate the executable atFUSIBILE_EXE_PATH
- Run post-processing (--prob_threshold 0.8 if using 3DCNNs):
python depthfusion.py --dense_folder TEST_DATA_FOLDER --fusibile_exe_path FUSIBILE_EXE_PATH --prob_threshold 0.3
- The final point cloud is stored in
TEST_DATA_FOLDER/points_mvsnet/consistencyCheck-TIME/final3d_model.ply
.
We observe that depthfusion.py
produce similar but quantitatively worse result to our own implementation. For detailed differences, please refer to MVSNet paper and Galliani's paper. The point cloud for scan9
should look like:
![]() |
![]() |
---|---|
point cloud result | ground truth point cloud |
The following steps are required to reproduce depth map/point cloud results:
- Generate R/MVSNet inputs from SfM outputs.You can use our preprocessed testing data in the download section. (provided)
- Run R/MVSNet testing script to generate depth maps for all views (provided)
- Run R/MVSNet validation script to generate depth map validation results. (provided)
- Apply variational depth map refinement for all views (optional, not provided)
- Apply depth map filter and fusion to generate point cloud results (partially provided via fusibile)
R-MVSNet point cloud results with full post-processing are also provided: DTU evaluation point clouds
Each project folder should contain the following
.
├── images
│ ├── 00000000.jpg
│ ├── 00000001.jpg
│ └── ...
├── cams
│ ├── 00000000_cam.txt
│ ├── 00000001_cam.txt
│ └── ...
└── pair.txt
If you want to apply R/MVSNet to your own data, please structure your data into such a folder. We also provide a simple script colmap2mvsnet.py
to convert COLMAP SfM result to R/MVSNet input.
All image files are stored in the images
folder. We index each image using an 8 digit number starting from 00000000
. The following camera and output files use the same indexes as well.
The camera parameter of one image is stored in a cam.txt
file. The text file contains the camera extrinsic E = [R|t]
, intrinsic K
and the depth range:
extrinsic
E00 E01 E02 E03
E10 E11 E12 E13
E20 E21 E22 E23
E30 E31 E32 E33
intrinsic
K00 K01 K02
K10 K11 K12
K20 K21 K22
DEPTH_MIN DEPTH_INTERVAL (DEPTH_NUM DEPTH_MAX)
Note that the depth range and depth resolution are determined by the minimum depth DEPTH_MIN
, the interval between two depth samples DEPTH_INTERVAL
, and also the depth sample number DEPTH_NUM
(or max_d
in the training/testing scripts if DEPTH_NUM
is not provided). We also left the interval_scale
for controlling the depth resolution. The maximum depth is then computed as:
DEPTH_MAX = DEPTH_MIN + (interval_scale * DEPTH_INTERVAL) * (max_d - 1)
We store the view selection result in the pair.txt
. For each reference image, we calculate its view selection scores with each of the other views, and store the 10 best views in the pair.txt file:
TOTAL_IMAGE_NUM
IMAGE_ID0 # index of reference image 0
10 ID0 SCORE0 ID1 SCORE1 ... # 10 best source images for reference image 0
IMAGE_ID1 # index of reference image 1
10 ID0 SCORE0 ID1 SCORE1 ... # 10 best source images for reference image 1
...
We provide a script to convert COLMAP SfM result to R/MVSNet input. After recovering SfM result and undistorting all images, COLMAP should generate a dense folder COLMAP/dense/
containing an undistorted image folder COLMAP/dense/images/
and an undistorted camera folder COLMAP/dense/sparse/
. Then, you can apply the following script to generate the R/MVSNet input:
python colmap2mvsnet.py --dense_folder COLMAP/dense
The depth sample number will be automatically computed using the inverse depth setting. If you want to generate the MVSNet input with a fixed depth sample number (e.g., 256), you could specified the depth number via --max_d 256
.
The test.py
script will create a depths_mvsnet
folder to store the running results, including the depth maps, probability maps, scaled/cropped images and the corresponding cameras. The depth and probability maps are stored in .pfm
format. We provide the python IO for pfm files in the preprocess.py
script, and for the c++ IO, we refer users to the Cimg library. To inspect the pfm format results, you can simply type python visualize.py .pfm
.
- Update BlendedMVG interface
- Pretrained models on BlendedMVS, DTU and ETH3D (trained for 150000 iterations)
- Update instructions in
README.md
- Training with BlendedMVS dataset
- Add
validate.py
script for depth map validation - Add
photometric_augmentation.py
script for online augmentation durig training
- Add
colmap2mvsnet.py
script to convert COLMAP SfM result to MVSNet input, including depth range estimation and view selection
- Add Baiduyun (code: s2v2) link for mainland China users
- Add R-MVSNet point clouds of DTU evaluation set
- Add "Reproduce Benchmarking Results" section
- MVSNet/R-MVSNet training/testing scripts
- MVSNet and R-MVSNet models (trained for 100000 iterations)
- Implement R-MVSNet and GRU regularization
- Network change: enable scale and center in batch normalization
- Network change: replace UniNet with 2D UNet
- Network change: use group normalization in R-MVSNet
- Use
tf.contrib.image.transform
for differentiable homography warping. Reconstruction is now x2 faster!