DTU-Evaluation

Matlab code for DTU dataset evaluation.

Environment

Matlab R2018a with statistics and machine learning toolbox.

Data preparation

For quantitative evaluation on DTU dataset, download SampleSet and Points. Unzip them and place Points folder in SampleSet/MVS Data/. The structure looks like:

SampleSet
├──MVS Data
      └──Points

Parameter settings

In BaseEvalMain_web.m, set parameters as following steps:

1. Set dataPath as path to standard point clouds like ${YOUR_ROOT_DIR}/SampleSet/MVS Data/.
2. Set plyPath as the directory that stores the reconstructed point clouds like ${YOUR_ROOT_DIR}/eval.
3. Set resultsPath as the directory to store the evaluation results(.mat file) like ${YOUR_ROOT_DIR}/results.
4. Set method_string as the prefix of the point cloud file. For example, if the file name of a reconstructed point cloud is final3d_model001_l3.ply, the prefix is final3d_model. The 001 means scan1 and l3 implies lighting condition.
5. Then run BaseEvalMain_web.m in matlab.