Self-supervised neural audio-visual sound source localization via probabilistic spatial modeling

Yoshiki Masuyama, Yoshiaki Bando, Kohei Yatabe, Yoko Sasaki, Masaki Onishi, Yasuhiro Oikawa

研究成果: Conference contribution

1 被引用数 (Scopus)

抄録

Detecting sound source objects within visual observation is important for autonomous robots to comprehend surrounding environments. Since sounding objects have a large variety with different appearances in our living environments, labeling all sounding objects is impossible in practice. This calls for self-supervised learning which does not require manual labeling. Most of conventional self-supervised learning uses monaural audio signals and images and cannot distinguish sound source objects having similar appearances due to poor spatial information in audio signals. To solve this problem, this paper presents a self-supervised training method using 360 images and multichannel audio signals. By incorporating with the spatial information in multichannel audio signals, our method trains deep neural networks (DNNs) to distinguish multiple sound source objects. Our system for localizing sound source objects in the image is composed of audio and visual DNNs. The visual DNN is trained to localize sound source candidates within an input image. The audio DNN verifies whether each candidate actually produces sound or not. These DNNs are jointly trained in a self-supervised manner based on a probabilistic spatial audio model. Experimental results with simulated data showed that the DNNs trained by our method localized multiple speakers. We also demonstrate that the visual DNN detected objects including talking visitors and specific exhibits from real data recorded in a science museum.

本文言語English
ホスト出版物のタイトル2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2020
出版社Institute of Electrical and Electronics Engineers Inc.
ページ4848-4854
ページ数7
ISBN(電子版)9781728162126
DOI
出版ステータスPublished - 2020 10 24
イベント2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2020 - Las Vegas, United States
継続期間: 2020 10 242021 1 24

出版物シリーズ

名前IEEE International Conference on Intelligent Robots and Systems
ISSN(印刷版)2153-0858
ISSN(電子版)2153-0866

Conference

Conference2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2020
国/地域United States
CityLas Vegas
Period20/10/2421/1/24

ASJC Scopus subject areas

  • 制御およびシステム工学
  • ソフトウェア
  • コンピュータ ビジョンおよびパターン認識
  • コンピュータ サイエンスの応用

フィンガープリント

「Self-supervised neural audio-visual sound source localization via probabilistic spatial modeling」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル