Dataset Augmentation and Dimensionality Reduction of Pinna-Related Transfer Functions

Guezenoc C, Séguier R. Dataset Augmentation and Dimensionality Reduction of Pinna-Related Transfer Functions. In: AES Convention 148. Audio Engineering Society; 2020. Paper 10337. Available from: https://aes.org/publications/elibrary-page/?id=20754

@inproceedings{Guezenoc2020_20754,
  author = {Guezenoc, Corentin and Séguier, Renaud},
  title = {{Dataset Augmentation and Dimensionality Reduction of Pinna-Related Transfer Functions}},
  booktitle = {AES Convention 148},
  note = {Paper 10337},
  year = {2020},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=20754}
}

TY  - CPAPER
TI  - Dataset Augmentation and Dimensionality Reduction of Pinna-Related Transfer Functions
AU  - Guezenoc, Corentin
AU  - Séguier, Renaud
T2  - AES Convention 148
M1  - Paper 10337
PY  - 2020
DA  - 2020/05/06
UR  - https://aes.org/publications/elibrary-page/?id=20754
PB  - Audio Engineering Society
LA  - en
AB  - Ef?cient modeling of the inter-individual variations of head-related transfer functions (HRTF) is a key matter to the individualization of binaural synthesis. In previous work, we augmented a dataset of 119 pairs of ear shapes and pinna-related transfer functions (PRTFs), thus creating a wide dataset of 1005 ear shapes and PRTFs generated by random ear drawings (WiDESPREaD) and acoustical simulations. In this article, we investigate the dimensionality reduction capacity of two principal component analysis (PCA) models of magnitude PRTFs, trained on WiDESPREaD and on the original dataset, respectively. We ?nd that the model trained on the WiDESPREaD dataset performs best, regardless of the number of retained principal components.
ER  -