Virtual LiDAR error models in point cloud compression

Repository for my Master thesis "Virtual LiDAR error models in point cloud compression".

Contents

The most important parts of the repository are listed below.

noise_distributions.ipynb

This notebook accumulates the analysis of the errors from the real and virtual scans.

pccarena_results.ipynb

This notebook accumulates the analysis of the results from the PCC Arena benchmarks. The input to this analysis is the summary.csv file which contains the measured results.

blend.empty

This file is the basis empty scene file used for all the scanning procedures in BlenSor.

scan_for_error.py

This script performs the virtual error measurements for a given error model with a floor and wall placed in the scene.

error_pipeline.py

Wraps the scan_for_error.py and adds some filesystem scaffolding.

s3dis_can.py

This script performs BlenSor scans in a given mesh file at given locations.

s3dis_pipeline.py

Wraps the s3dis_scan.py and adds some filesystem scaffolding as well as the cropping by bounding box operation implemented in crop_by_bounding_box.py.

exploration_scan_script.py

This script was used for the construction of the scanning pipeline and other explorations.

evd_reader.py

Contains classes to read BlenSor's .evd file and output a .txt or .pcd file.

Other work

Apart from the work in this repository, modifications to Indoor reconstruction++ were made as a Pull Request to the original repository, available here. The work done on BlenSor is available in a fork of the original repository found here.