Mesh2scattering

------------------
* initial commit

Mesh2scattering Developers
==========================

Mesh2scattering is currently maintained and developed by
Anne Heimes

**Institutions**

* Institute for Hearing Technology and Acoustics, RWTH Aachen University, Germany

Mesh2HRTF Developers
====================

Mesh2HRTF is currently maintained and developed by
Piotr Majdak (Conceptualization),
Fabian Brinkmann (Python & Matlab API, Blender Export, Testing, Documentation),
Wolfang Kreuzer (NumCalc, Documentation),
Katharina Pollack (Matlab API, Documentation)

Contributors
============

The following persons contributed to Mesh2HRTF, named in reverse chronological
order

* Jeffrey Thomsen (2022): Testing and documentation
* Sergejs Dombrovskis (2022): Initial NumCalc manager version, documentation and tutorials
* Timon Palm (2021): Hybrid mesh grading tool
* Sebastian Koch (2021): Hybrid meh grading tool
* Junaid Khan (2020): Bugfixes and restructuring
* Oliver Weissbarth (2020): Update of the OpenFlipper mesh grading plug-in
* Slim Ghorbal (2019): Improved Blender export
* Robert Pelzer (2018): Blender AddOns for head centering and material assignment
* Harald Ziegelwanger (2016-2018): Initial development of Mesh2HRTF
* Z. S. Chen (2016): Initial development of NumCalc
* Michael Kalcher
* Johan Pauwels

**Institutions**

* Austrian Research Institute, Austian Academy of Sciences, Vienna
* Audio Communication Group, Technical University of Berlin
* Computer Graphics Group, Technical University of Berlin
* University of Applied Sciences, Technikum Wien
* Imperial College London, UK
* Royal Institute of Technology, Stockholm, Sweden
* Mimi Hearing Technologies, Berlin, Germany

-------
* Project export from Blender (handled by the Blender plugin `Mesh2Input/mesh2input.py`)
* Upgraded to Blender >= 2.80.0
* Better organization and modularization of the Blender plugin
* Re-design the parameters, appearance, and in-app documentation of the export menu to be less prone to erroneous input
* The source type is now defined by a single drop down menu
* Source properties (point source position and plane wave direction) are now obtained from Objects in the Blender Scene and do not have to be entered manually anymore.
* Added an option for referencing the HRTF to a source in the origin of coordinates according to the HRTF definition
* Referencing is now also supported for point sources
* Add an option to calculate HRIRs from single sided HRTF spectra
* More flexible selection of the frequencies that are simulated
* Multiple evaluation grids can be entered in a single text field
* Added handling for custom definitions of materials (boundary conditions)
* Moved options for parallelization from the export menu to NumCalc (see below)
* Removed undocumented and unfinished options fro near-field calculation and frequency dependent meshes
* Boundary conditions can now be frequency dependent
* Files containing custom evaluation grids and material data can be located outside the Mesh2HRTF repository
* NumCalc (contained in NumCalc/Source)
* Added NumCalc manager for automatic parallelization of frequency steps
* Bugfix: when NC.inp does not contain an "END" of file, NumCalc now throws one error and aborts programme execution (was infinite loop resulting in large output files)
* Introduced command line parameters `istart` and `iend` to select a range of frequencies for simulation and to ease parallelization
* Introduce command line parameter `nitermax` to control the maximum number of iterations
* Introduce command line parameter `estimate_ram` for an a priori estimate of the RAM consumption per frequency step
* Reduce the default number of maximum iterations to 250
* Minor bug fixes and stability improvements
* Added a Python API for processing NumCalc output and save HRTF and HRIR data
* Installable via pip
* Full online documentation
* Added function to generate a project report and notify in case of issues and/or errors that occurred during the NumCalc simulation
* Added flexible plot function for quick inspection of the results
* Added Python tools to read and generate custom evaluation grids
* Added function to merge results from multiple sources (e.g. left and right ear) into a single SOFA file
* Added function to write boundary conditions to material files
* Improved structure of the output data (Sofa files, project reports, exports, plots)
* Data is now stored in a separate folder `Output2HRTF`
* Data is named according to the evaluation girds and object meshes
* Data for multiple evaluation grids is now stored in separate files
* Frequencies in SOFA files now contain decimal values
* Added testing
* All three parts of Mesh2HRTF (Project export, NumCalc, and Output2HRTF) are tested using pytest to improve and monitor the code quality
* The Matlab/Octave API is not tested. New users are recommended to use the Python API
* General
* Unified names of functions across programming languages
* Updated project wiki and moved to github

------
* this will be the last release supporting Blender < 2.8
* new directory structure

------
* big fix and improvements in PreProcessing/MeshGrading (ticket 25, commit r38)
* bug fix in ExportMesh2HRTF.py (Tickets 13, 15, and 23 on sourcefourge)
* fixed a bug Output2HRTF_Main.m. Now the low frequency HRTF should at 0 dB if using reciprocal simulation and if setting reference = true in Output2HRTF.m, which is auto-generated when exporting from blender.

------
* bug fix in NumCalc

------
* New directory structure
* Pascal-case naming of the files
* Small bugfixes in the scripts