Wissenschaftler
Fachbereich Mathematik
Frame Theory and its Implementation
Sprecher des Teams Maschinelles Lernen

Tel. +43 1 51581-2532
Email: nicki.holighaus(at)oeaw.ac.at

Wissenschaftliche IDs:
ORCID: 0000-0003-3837-2865
Google Scholar: Nicki Holighaus
ResearchGate: researchgate.net/profile/Nicki_Holighaus

 

Bildung

Nicki Holighaus studierte Mathematik und theoretische Informatik an der Justus–Liebig–Universität zu Gießen, Deutschland. Nach dem Abschluss im Jahr 2010 begann er das Dokrotratsstudium an der Universität Wien, Österreich, dass er im Oktober 2013 mit der Verteidigung seiner Doktorarbeit "Theory and implementation of adaptive time-frequency transforms” erfolgreich abschloss. Während seines Studiums an der Universität Wien arbeitete er dort als Forschungsassistent in der Numerical Harmonic Analysis Group (NuHAG). 

Seit August 2012 ist er Mitglied der Gruppe "Mathematics and Signal Processing in Acoustics" des Instituts für Schallforschung. Dort arbeitet er an theoretischen und angewandten Aspekten adaptierter Zeit-Frequenz Darstellungen und anderer Frames.

Derzeitige Forschung

Seine Forschungsinteressen konzentrieren sich auf die Verwendung von Zeit-Frequenz-Methoden für Signalverarbeitung. Insbesondere forscht er in den Bereichen Zeit-Frequenz-Analyse, mathematische Theorie und Design adaptiver und adaptierter Zeit-Frequenz-Darstellungen, Zeit-Frequency-Verarbeitung in der Akustik und Verwendung von Zeit-Frequenz-Darstellungen in maschinellem Lernen für akustische Signalverarbeitung.

Aktuelle Forschungsprojekte: MERLIN

Aktuelle Themen:

  • Theory and application of warped time-frequency representations 
  • Function spaces and discretization for structured continuous frames
  • Structure of time-frequency phase
  • Signal processing with time-frequency phase
  • Deep learning with time-frequency features
  • Neural audio generation
  • Audio inpainting with generative neural networks
  • Time-frequency processing and perception

Publikationen

  • Prediction of parameters of a pinna model from synthetic geometries using a vision transformer. / Pausch, Florian; Perfler, Felix; Holighaus, Nicki et al.
    in: Journal of the Acoustical Society of America, Jahrgang 159, Nr. 6, 18.06.2026, S. 5578-5598.
    DOIRISENWBIB

    The acquisition of the human pinna geometry requires elaborate equipment for accurate results. Even then, the results are often corrupted by measurement artifacts. We introduce Mesh2PPM, a framework facilitating the generation of a personalized and artifact-free pinna mesh. Mesh2PPM predicts the parameters of a parametric pinna model based on cubic Bézier curves (BezierPPM) from a pinna mesh via a vision transformer. We evaluated Mesh2PPM with multi-view renderings of synthetic pinna geometries, providing additional depth information, varying the grids of camera views, and jittering the camera views. While added depth information had no practically relevant effect, a grid with 3 ×3 camera views facilitated the lowest overall prediction errors and best counteracted the detrimental effects of jitter. For this grid, with and without jitter, the median Pompeiu-Hausdorff distances were 1.98 mm and 1.34 mm, respectively, and the root mean square distances were 0.92 mm and 0.52 mm. A refined analysis targeting the perceptually most important pinna regions for sound localization showed that multi-view information particularly improved the prediction of BezierPPM parameters describing the cavum-conchae region. The accuracy achieved indicates the suitability of Mesh2PPM to retrieve BezierPPM parameters from pinna meshes.

  • On the crossband filter representations of LTI systems: Algorithms and inversion. / Matsuyama, Naoya; Holighaus, Nicki; Yamada, Koki et al.
    in: Acoustical Science and Technology, Jahrgang 47, Nr. 1, 01.01.2026, S. 13-23.
    DOIRISENWBIB
  • Discretization of Continuous Frames by Quasi-Monte Carlo Methods. / Zimmermann, Jan; Klotz, Andreas; Holighaus, Nicki.
    2025 International Conference on Sampling Theory and Applications (SampTA). IEEE, 2025.
    DOIRISENWBIB
  • ISAC: An Invertible and Stable Auditory Filter Bank with Customizable Kernels for ML Integration. / Haider, Daniel; Perfler, Felix; Balazs, Peter et al.
    2025 International Conference on Sampling Theory and Applications (SampTA). IEEE, 2025.
    DOIRISENWBIB
  • Mesh2PPM - Automatic Parametrization of the BezierPPM: Entire Pinna. / Pausch, Florian; Perfler, Felix; Holighaus, Nicki et al.
    AES Europe 2025 Conference Proceedings. Audio Engineering Society, 2025. 336.
    WebseiteRISENWBIB
  • Parametric model of the human pinna based on Bézier curves and concave deformations. / Perfler, F; Pausch, F; Pollack, K et al.
    in: Computers in Biology and Medicine, Jahrgang 188, 109817, 01.04.2025.
    DOIRISENWBIB
  • Robust Multicomponent Tracking of Ultrasonic Vocalizations. / Abbasi, R; Holighaus, N; Balazs, P et al.
    ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Hyderabad, 2025. S. 1-5.
    DOIRISENWBIB
  • Coorbit theory of warped time-frequency systems in R^d. / Holighaus, N; Voigtlaender, F.
    in: Journal of Fourier Analysis and Applications, Jahrgang 30, 25.10.2024, S. Nr. 62.
    DOIRISENWBIB
  • LTFATPY: towards making a wide range of time-frequency representations available in Python. / Hollomey, C; Holighaus, N.
    Proceedings of the 27th International Conference on Digital Audio Effects. Guildford, UK, 2024. S. 479-482.
    RISENWBIB
  • Time-frequency analysis on flat tori and Gabor frames in finite dimensions. / Abreu, L; Balazs, P; Holighaus, N et al.
    in: Applied and Computational Harmonic Analysis, Jahrgang 69, 01.03.2024, S. 101622.
    DOIRISENWBIB
  • Comparison of deep-neural-network architectures for the prediction of head-related transfer functions using a parametric pinna model. / Pausch, F; Perfler, F; Holighaus, N et al.
    Proceedings of the Forum Acusticum 2023. Turin, 2023.
    DOIRISENWBIB
  • Rotated time-frequency lattices are sets of stable sampling for continuous wavelet systems. / Holighaus, N; Koliander, G.
    14th International Conference on Sampling Theory and Applications. Yale, 2023.
    DownloadRISENWBIB
  • Grid-Based Decimation for Wavelet Transforms With Stably Invertible Implementation. / Holighaus, N; Koliander, G; Hollomey, C et al.
    in: IEEE/ACM Transactions on Audio Speech and Language Processing, Jahrgang 31, 13.01.2023, S. 789-801.
    DOIRISENWBIB
  • Grid-like wavelet sampling for audio processing applications. / Hollomey, C; Holighaus, N; Koliander, G.
    Proceedings: A16, Numerical, Computational and Theoretical Acoustics, ICA 2022. Gyeongju, 2022. S. 208-214.
    RISENWBIB
  • Music signal analysis in the Large Time Frequency Analysis Toolbox. / Hollomey, C; Holighaus, N; Balazs, P.
    Proceedings: A07, Musical Acoustics, ICA 2022. Gyeongju, 2022. S. 98-104.
    RISENWBIB
  • Audio Inpainting. / Balazs, P; Tauböck, G; Rajbamshi, S et al.
    Proceedings: A16, Numerical, Computational and Theoretical Acoustics, ICA 2022. Gyeongju, 2022. S. 186-189.
    RISENWBIB
  • Gutachten für DAFx2022_07_2022. / Holighaus, Nicki.
    2022.
    RISENWBIB
  • Fast Matching Pursuit with Multi-Gabor Dictionaries. / Průša, Z; Holighaus, N; Balazs, P.
    in: Transactions on Mathematical Software, Jahrgang 47, 01.12.2021, S. 1-20.
    DOIRISENWBIB
  • Zeit-Frequenz Darstellungen und Deep Learning. / Haider, D; Balazs, P; Holighaus, N et al.
    DAGA 2021, Jahrestagung für Akustik. Vienna, 2021.
    RISENWBIB
  • Audio Inpainting via L1-Minimization and Dictionary Learning. / Rajbamshi, S; Tauböck, G; Holighaus, N et al.
    European Signal Processing Conference (EUSIPCO 2021). 2021.
    RISENWBIB