Machine learning based prediction for Personalized Head Related Transfer Functions based on video capture

Javeri N, Dutta PB, Sunder K, Jain K. Machine learning based prediction for Personalized Head Related Transfer Functions based on video capture. In: AES Conference: AES 2022 International Audio for Virtual and Augmented Reality Conference. Audio Engineering Society; 2022. Paper 26. Available from: https://aes.org/publications/elibrary-page/?id=21856

@inproceedings{Javeri2022_21856,
  author = {Javeri, Nikhil and Dutta, Prabal Bijoy and Sunder, Kaushik and Jain, Kapil},
  title = {{Machine learning based prediction for Personalized Head Related Transfer Functions based on video capture}},
  booktitle = {AES Conference: AES 2022 International Audio for Virtual and Augmented Reality Conference},
  note = {Paper 26},
  year = {2022},
  month = aug,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=21856}
}

TY  - CPAPER
TI  - Machine learning based prediction for Personalized Head Related Transfer Functions based on video capture
AU  - Javeri, Nikhil
AU  - Dutta, Prabal Bijoy
AU  - Sunder, Kaushik
AU  - Jain, Kapil
T2  - AES Conference: AES 2022 International Audio for Virtual and Augmented Reality Conference
M1  - Paper 26
PY  - 2022
DA  - 2022/08/06
UR  - https://aes.org/publications/elibrary-page/?id=21856
PB  - Audio Engineering Society
LA  - en
AB  - Over the past decade, audio for extended reality has become critical to deliver a truly immersive sound experience. With headphones being a popular medium for playback, binaural audio is one of the most convenient formats to deliver accurate spatial audio. Personalized Head-related Transfer Functions (HRTFs) are an integral component of binaural audio that determines the quality of the spatial audio experience. In this paper, we present a pilot research that predicts personalized HRTFs based on 2D images or a video capture. We explore different components in this process including the 3D reconstruction of an ear based on 2D images or video followed by the HRTF estimation using HRTF prediction using Neural Networks.
ER  -