Online Multimodal Speaker Detection for Humanoid Robots

Jordi Sanchez-Riera (1) , Xavier Alameda-Pineda (1) , Johannes Wienke (2) , Antoine Deleforge (1) , Soraya Arias (3) , Jan Cech (1) , Sebastian Wrede (4) , Radu Horaud (1)

1 PERCEPTION - Interpretation and Modelling of Images and Videos
2 Universität Bielefeld = Bielefeld University
3 SED [Grenoble] - Service Expérimentation et Développement
4 CoR-Lab - Research Institute for Cognition and Robotics

In this paper we address the problem of audio-visual speaker detection. We introduce an online system working on the humanoid robot NAO. The scene is perceived with two cameras and two microphones. A multimodal Gaussian mixture model (mGMM) fuses the information extracted from the auditory and visual sensors and detects the most probable audio-visual object, e.g., a person emitting a sound, in the 3D space. The system is implemented on top of a platform-independent middleware and it is able to process the information online (17Hz). A detailed description of the system and its implementation are provided, with special emphasis on the online processing issues and the proposed solutions. Experimental validation, performed with five different scenarios, show that that the proposed method opens the door to robust human-robot interaction scenarios.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Jordi Sanchez-Riera, Xavier Alameda-Pineda, Johannes Wienke, Antoine Deleforge, Soraya Arias, et al.. Online Multimodal Speaker Detection for Humanoid Robots. Humanoids 2012 - IEEE International Conference on Humanoid Robotics, Nov 2012, Osaka, Japan. pp.126-133, ⟨10.1109/HUMANOIDS.2012.6651509⟩. ⟨hal-00768764⟩