Calculation of listener-specific head-related transfer functions: Effect of mesh quality

Audibility of dispersion error in room acoustic finite-difference time-domain simulation in the presence of a single early reflection

“…In such case, a 3-D model of the subject's head need to be constructed using scanning techniques. Naturally, the quality of the mesh has been identified as an important factor in determining the quality of the simulated HRTFs [147].…”

Numerical calculation of listener-specific head-related transfer functions and sound localization: Microphone model and mesh discretization

“…However, the laser scanner was not able to scan hairy surfaces. Thus, the subjects wore a custom tailored cap with holes for the ears (Ziegelwanger et al , 2013). In a post-processing procedure, a 3D triangular polygon mesh was extracted from a volumetric representation with approximately 0.5 mm voxel size.…”

“…Such a scanner is not limited in the amount of radiation by medical concerns and is capable to deliver a detailed representation of the pinna geometry. To prevent the subjects from the harmful x-ray radiation, silicone impressions of subjects’ pinnae and surrounding area were made (viscosity: 2200 MPa·s, shore hardness: A 22, for more details on the silicon impressions see Ziegelwanger et al , 2013). The impressions were then scanned using 1000 x-ray images spaced in 0.36° intervals (tube settings: 120 kV, 260 μ A, average of three images with 333-ms exposure time each).…”

“…Note that the pinnae meshes from the CT scans were not manipulated in this procedure. The final reference meshes consisted of approximately 2.5 × 10 6 elements and had an average edge length (AEL) and a maximum edge length of 0.47 and 0.77 mm, respectively (for more details on the geometry acquisition, see Reichinger et al , 2013; Ziegelwanger et al , 2013).…”

“…Acoustically measured HRTFs were validated in sound-localization experiments (Bronkhorst, 1995; Middlebrooks, 1999a) by comparing listener-specific localization-performance parameters for real and virtual sound sources. Recently, various models have been developed to estimate such parameters (Baumgartner et al , 2013; Ziegelwanger and Majdak, 2014) and they can be used to quantify the similarity of HRTF sets on a perceptual level (Ziegelwanger et al , 2013).…”