Epipolar geometry based sound localization and extraction for humanoid audition

Kazuhiro Nakadai, Hiroshi G. Okuno, Hiroaki Kitano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

39 Citations (Scopus)

Abstract

Sound localization for a robot or an embedded system is usually solved by using Interaural Phase Difference (IPD) and Interaural Intensity Difference (IID). These values are calculated by using Head-Related Transfer Function (HRTF). However, HRTF depends on the shape of head and also changes as environments changes. Therefore, sound localization without HRTF is needed for real-world applications. In this paper, we present a new sound localization method based on auditory epipolar geometry with motion control. Auditory epipolar geometry is an extension of epipolar geometry in stereo vision to audition, and auditory and visual epipolar geometry can share the sound source direction. The key idea is to exploit additional inputs obtained by motor control in order to compensate damages in the IPD and IID caused by reverberation of the room and the body of a robot. The proposed system can localize and extract simultaneous two sound sources in a real-world room.

Original languageEnglish
Title of host publicationIEEE International Conference on Intelligent Robots and Systems
Pages1395-1401
Number of pages7
Volume3
Publication statusPublished - 2001
Externally publishedYes
Event2001 IEEE/RSJ International Conference on Intelligent Robots and Systems - Maui, HI
Duration: 2001 Oct 292001 Nov 3

Other

Other2001 IEEE/RSJ International Conference on Intelligent Robots and Systems
CityMaui, HI
Period01/10/2901/11/3

Fingerprint

Audition
Acoustic waves
Geometry
Transfer functions
Robots
Stereo vision
Reverberation
Motion control
Embedded systems

Keywords

  • Active audition
  • Humanoid
  • Localization
  • Sensor fusion

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Nakadai, K., Okuno, H. G., & Kitano, H. (2001). Epipolar geometry based sound localization and extraction for humanoid audition. In IEEE International Conference on Intelligent Robots and Systems (Vol. 3, pp. 1395-1401)

Epipolar geometry based sound localization and extraction for humanoid audition. / Nakadai, Kazuhiro; Okuno, Hiroshi G.; Kitano, Hiroaki.

IEEE International Conference on Intelligent Robots and Systems. Vol. 3 2001. p. 1395-1401.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nakadai, K, Okuno, HG & Kitano, H 2001, Epipolar geometry based sound localization and extraction for humanoid audition. in IEEE International Conference on Intelligent Robots and Systems. vol. 3, pp. 1395-1401, 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems, Maui, HI, 01/10/29.
Nakadai K, Okuno HG, Kitano H. Epipolar geometry based sound localization and extraction for humanoid audition. In IEEE International Conference on Intelligent Robots and Systems. Vol. 3. 2001. p. 1395-1401
Nakadai, Kazuhiro ; Okuno, Hiroshi G. ; Kitano, Hiroaki. / Epipolar geometry based sound localization and extraction for humanoid audition. IEEE International Conference on Intelligent Robots and Systems. Vol. 3 2001. pp. 1395-1401
@inproceedings{e080c7b51fbb413ab1596315fde0945c,
title = "Epipolar geometry based sound localization and extraction for humanoid audition",
abstract = "Sound localization for a robot or an embedded system is usually solved by using Interaural Phase Difference (IPD) and Interaural Intensity Difference (IID). These values are calculated by using Head-Related Transfer Function (HRTF). However, HRTF depends on the shape of head and also changes as environments changes. Therefore, sound localization without HRTF is needed for real-world applications. In this paper, we present a new sound localization method based on auditory epipolar geometry with motion control. Auditory epipolar geometry is an extension of epipolar geometry in stereo vision to audition, and auditory and visual epipolar geometry can share the sound source direction. The key idea is to exploit additional inputs obtained by motor control in order to compensate damages in the IPD and IID caused by reverberation of the room and the body of a robot. The proposed system can localize and extract simultaneous two sound sources in a real-world room.",
keywords = "Active audition, Humanoid, Localization, Sensor fusion",
author = "Kazuhiro Nakadai and Okuno, {Hiroshi G.} and Hiroaki Kitano",
year = "2001",
language = "English",
volume = "3",
pages = "1395--1401",
booktitle = "IEEE International Conference on Intelligent Robots and Systems",

}

TY - GEN

T1 - Epipolar geometry based sound localization and extraction for humanoid audition

AU - Nakadai, Kazuhiro

AU - Okuno, Hiroshi G.

AU - Kitano, Hiroaki

PY - 2001

Y1 - 2001

N2 - Sound localization for a robot or an embedded system is usually solved by using Interaural Phase Difference (IPD) and Interaural Intensity Difference (IID). These values are calculated by using Head-Related Transfer Function (HRTF). However, HRTF depends on the shape of head and also changes as environments changes. Therefore, sound localization without HRTF is needed for real-world applications. In this paper, we present a new sound localization method based on auditory epipolar geometry with motion control. Auditory epipolar geometry is an extension of epipolar geometry in stereo vision to audition, and auditory and visual epipolar geometry can share the sound source direction. The key idea is to exploit additional inputs obtained by motor control in order to compensate damages in the IPD and IID caused by reverberation of the room and the body of a robot. The proposed system can localize and extract simultaneous two sound sources in a real-world room.

AB - Sound localization for a robot or an embedded system is usually solved by using Interaural Phase Difference (IPD) and Interaural Intensity Difference (IID). These values are calculated by using Head-Related Transfer Function (HRTF). However, HRTF depends on the shape of head and also changes as environments changes. Therefore, sound localization without HRTF is needed for real-world applications. In this paper, we present a new sound localization method based on auditory epipolar geometry with motion control. Auditory epipolar geometry is an extension of epipolar geometry in stereo vision to audition, and auditory and visual epipolar geometry can share the sound source direction. The key idea is to exploit additional inputs obtained by motor control in order to compensate damages in the IPD and IID caused by reverberation of the room and the body of a robot. The proposed system can localize and extract simultaneous two sound sources in a real-world room.

KW - Active audition

KW - Humanoid

KW - Localization

KW - Sensor fusion

UR - http://www.scopus.com/inward/record.url?scp=0035558048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035558048&partnerID=8YFLogxK

M3 - Conference contribution

VL - 3

SP - 1395

EP - 1401

BT - IEEE International Conference on Intelligent Robots and Systems

ER -