A hybrid approach to structural modeling of individualized HRTFs

Riccardo Miccini; Simone Spagnol

doi:10.1109/VRW52623.2021.00022

A hybrid approach to structural modeling of individualized HRTFs

Riccardo Miccini^*, Simone Spagnol

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

12 Citations (Scopus)

Abstract

We present a hybrid approach to individualized head-related transfer function (HRTF) modeling which requires only 3 anthropometric measurements and an image of the pinna. A prediction algorithm based on variational autoencoders synthesizes a pinna-related response from the image, which is used to filter a measured head-andtorso response. The interaural time difference is then manipulated to match that of the HUTUBS dataset subject minimizing the predicted localization error. The results are evaluated using spectral distortion and an auditory localization model. While the latter is inconclusive regarding the efficacy of the structural model, the former metric shows promising results with encoding HRTFs. Index Terms: Hardware - Digital signal processing; Computing methodologies - Neural networks; Applied computing - Sound and music computing

Original language	English
Title of host publication	Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021
Number of pages	6
Publisher	IEEE (Institute of Electrical and Electronics Engineers)
Publication date	Mar 2021
Pages	80-85
Article number	9419096
ISBN (Electronic)	9780738113678
DOIs	https://doi.org/10.1109/VRW52623.2021.00022
Publication status	Published - Mar 2021
Event	2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021 - Virtual, Lisbon, Portugal Duration: 27 Mar 2021 → 3 Apr 2021

Conference

Conference	2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021
Country/Territory	Portugal
City	Virtual, Lisbon
Period	27/03/2021 → 03/04/2021

Bibliographical note

Publisher Copyright:
© 2021 IEEE.

Keywords

Digital signal processing
Neural networks
Sound and music computing

Access to Document

10.1109/VRW52623.2021.00022

https://repository.tudelft.nl/islandora/object/uuid:37bad4b1-09ef-4386-bb56-5e52c18caa3a/datastream/OBJ/download

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{ed9e6faffbe94f8591d343b5ce24d5a8,

title = "A hybrid approach to structural modeling of individualized HRTFs",

abstract = "We present a hybrid approach to individualized head-related transfer function (HRTF) modeling which requires only 3 anthropometric measurements and an image of the pinna. A prediction algorithm based on variational autoencoders synthesizes a pinna-related response from the image, which is used to filter a measured head-andtorso response. The interaural time difference is then manipulated to match that of the HUTUBS dataset subject minimizing the predicted localization error. The results are evaluated using spectral distortion and an auditory localization model. While the latter is inconclusive regarding the efficacy of the structural model, the former metric shows promising results with encoding HRTFs. Index Terms: Hardware - Digital signal processing; Computing methodologies - Neural networks; Applied computing - Sound and music computing",

keywords = "Digital signal processing, Neural networks, Sound and music computing",

author = "Riccardo Miccini and Simone Spagnol",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021 ; Conference date: 27-03-2021 Through 03-04-2021",

year = "2021",

month = mar,

doi = "10.1109/VRW52623.2021.00022",

language = "English",

pages = "80--85",

booktitle = "Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021",

publisher = "IEEE (Institute of Electrical and Electronics Engineers)",

address = "United States",

}

Miccini, R & Spagnol, S 2021, A hybrid approach to structural modeling of individualized HRTFs. in Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021., 9419096, IEEE (Institute of Electrical and Electronics Engineers), pp. 80-85, 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021, Virtual, Lisbon, Portugal, 27/03/2021. https://doi.org/10.1109/VRW52623.2021.00022

A hybrid approach to structural modeling of individualized HRTFs. / Miccini, Riccardo; Spagnol, Simone.
Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021. IEEE (Institute of Electrical and Electronics Engineers), 2021. p. 80-85 9419096.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - A hybrid approach to structural modeling of individualized HRTFs

AU - Miccini, Riccardo

AU - Spagnol, Simone

PY - 2021/3

Y1 - 2021/3

N2 - We present a hybrid approach to individualized head-related transfer function (HRTF) modeling which requires only 3 anthropometric measurements and an image of the pinna. A prediction algorithm based on variational autoencoders synthesizes a pinna-related response from the image, which is used to filter a measured head-andtorso response. The interaural time difference is then manipulated to match that of the HUTUBS dataset subject minimizing the predicted localization error. The results are evaluated using spectral distortion and an auditory localization model. While the latter is inconclusive regarding the efficacy of the structural model, the former metric shows promising results with encoding HRTFs. Index Terms: Hardware - Digital signal processing; Computing methodologies - Neural networks; Applied computing - Sound and music computing

AB - We present a hybrid approach to individualized head-related transfer function (HRTF) modeling which requires only 3 anthropometric measurements and an image of the pinna. A prediction algorithm based on variational autoencoders synthesizes a pinna-related response from the image, which is used to filter a measured head-andtorso response. The interaural time difference is then manipulated to match that of the HUTUBS dataset subject minimizing the predicted localization error. The results are evaluated using spectral distortion and an auditory localization model. While the latter is inconclusive regarding the efficacy of the structural model, the former metric shows promising results with encoding HRTFs. Index Terms: Hardware - Digital signal processing; Computing methodologies - Neural networks; Applied computing - Sound and music computing

KW - Digital signal processing

KW - Neural networks

KW - Sound and music computing

UR - http://www.scopus.com/inward/record.url?scp=85105967061&partnerID=8YFLogxK

U2 - 10.1109/VRW52623.2021.00022

DO - 10.1109/VRW52623.2021.00022

M3 - Article in proceeding

AN - SCOPUS:85105967061

SP - 80

EP - 85

BT - Proceedings - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021

PB - IEEE (Institute of Electrical and Electronics Engineers)

T2 - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2021

Y2 - 27 March 2021 through 3 April 2021

ER -

A hybrid approach to structural modeling of individualized HRTFs

Abstract

Conference

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this