Radu HORAUD
Author's posts
Nov 05
Acoustic Space Mapping: A Machine Learning Approach to Sound Source Separation and Localization, by Antoine Deleforge
Tuesday, November 26, 2013, 10:00 to 10:45, Grand Amphi, INRIA Montbonnot PhD public defense by Antoine Deleforge, Perception Team, INRIA Montbonnot Abstract: In this thesis, we address the long-studied problem of binaural (two microphones) sound source separation and localization through supervised learning. To achieve this, we develop a new paradigm referred as acoustic space mapping, …
Nov 03
Modelling via Mixtures of Skew Distributions, by Geoff McLachlan
Tuesday, November 26, 2013, 14:30 to 15:30, Grand Amphithéatre, INRIA Montbonnot Seminar by Geoff McLachlan, Department of Mathematics, University of Queensland, Australia Abstract. Non-normal mixture distributions have received increasing attention in recent years. Finite mixtures of multivariate skew symmetric distributions, in particular, the skew normal and skew $t$-mixture models, are emerging as a promising extension to …
Nov 01
Visual Uncertainty: A Bayesian Approach, by Simon Barthelmé
Wednesday, November 13, 2013, 14:00 to 15:00, room F107, INRIA Montbonnot Seminar by Simon Barthelmé, Université de Genève Abstract. Visual processing is fraught with uncertainty: the brain’s visual system must attempt to estimate physical properties despite missing information and noisy mechanisms. Sometimes high visual uncertainty translates into lack of confidence in our visual perception: we …
Oct 28
ERC Advanced Grant Awarded to Radu Horaud
The PERCEPTION team is pleased to announce that Radu Horaud was awarded an ERC Advanced Grant for his project “Vision and Hearing in Action” (VHIA). VHIA investigates the fundamentals of audio-visual perception for human-robot interaction. Read an interview (in French) about the concept of social robot that is being developed during the lifetime of the …
Oct 04
IEEE MMSP’13: Best Paper Award!
The article “Alignment of Binocular-Binaural Data Using a Moving Audio-Visual Target” received the “Best Paper Award” at the IEEE International Workshop on Multimedia Signal Processing (MMSP’13), Pula, Italy, September-October 2013. The paper is authored by Vasil Khalidov (IDIAP), and co-authored by Radu Horaud (INRIA) and Florence Forbes (INRIA). The paper addresses the problem of of aligning visual and auditory data …
Sep 25
TransforMesh: A mesh-based solution to surface evolution.
TransforMesh is a C++ function library that provides a mesh-based solution to surface evolution. It takes as input a proper mesh together with a displacement field and it outputs a proper deformed mesh. In order to ensure consistent mesh deformations, it robustly handles topology changes and removes self intersections.. Among others, TransforMesh has been …
Sep 25
3D keypoints (MeshDOG) and local descriptors (MeshHOG) for uniformly triangulated meshes
This is a C++ implementation of a 3D feature detector (MeshDOG) and a 3D feature descriptor (MeshHOG) for uniformly triangulated meshes, invariant to changes in rotation, translation, and scale. Both the detector and the descriptor are able to capture the local geometric and/or photometric properties in a succinct fashion. Moreover, they are defined generically for …
Sep 25
The Future of Television, by Josep R. Casas
Tuesday, October 15, 2013, 2:00 to 3:00, room C207, INRIA Montbonnot by Josep Ramon Casas, Technical University of Catalonia (UPC), Barcelona Abstract. Television, a telecommunication medium for the transmission of moving images and sound (cf.wikipedia). It substantiates the “far sight” concept of its mixed Greek-Latin etymology. But this definition may be misleading. Are the current technological solutions …
Sep 23
Master projects for the academic year 2013-2104
The PERCEPTION team (Interpretation and Modeling of Images and Sounds) offers a number of Master projects for the academic year 2013-2014. Each project has a duration of 3 to 6 months starting any time in 2014. The selected students will work at INRIA Montbonnot (near Grenoble) and they will receive a stipend: students enrolled in …
Sep 23
(closed) Master Project: Sound-source localization based on audio-visual mapping learning
Deadline for sending applications: 30 November 2013. Project proposed by Radu Horaud and Antoine Deleforge. We propose to address the interesting and challenging problem of sound-source localization. A sound source is generally localized using the time-difference of arrival (TDOA) between pairs of microphones. In general, the microphones are arranged in a linear or circular array …
