Photo of Xavier Alameda-Pineda

Xavier Alameda-Pineda

Research Director and Leader of the RobotLearn Team

Research Overview

My research is at the intersection of machine learning, computer vision, and audio processing, with a focus on multimodal perception and learning for social robots. I lead the RobotLearn team at Inria.

Awards

Projects

Anandabot

Partner · 2026–2030

None

muDialbot

Partner · 2021-2025 · ANR

Multi-party perceptually-active situated dialog for human-robot interaction

SPRING

Coordinator · 2020–2024 · H2020, 8.3 M€

H2020 project: Socially pertinent robots for gerontological healthcare

ML3RI

PI · 2020–2024 · ANR-JCJC & IDEX-IRS, 310 k€

ANR-JCJC: Multi-modal multi-person low-level learning for robot interaction

MIAI Chair

co-PI · 2019–2023 · ANR 3IA, 360 k€

ANR 3IA: Audio-visual machine perception and interaction for companion robots

PIMPE

PI · 2019–2023 · IDEX-ISP, 45 k€

IDEX-ISP: Physical complex Interactions and Multi-person Pose Estimation

MIDGen

co-PI · 2019–2023 · IDEX-IRS, 90 k€

IDEX-IRS: Multimodal Interaction Data Generation

Short Bio

Xavier Alameda-Pineda holds the position of Research Director at Inria and is the Leader of the RobotLearn Team. He obtained the M.Sc. (equivalent) in Mathematics in 2008, Telecommunications in 2009 from BarcelonaTech, and Computer Science in 2010 from Univ. Grenoble-Alpes (UGA). He then worked towards his Ph.D. in Mathematics and Computer Science, and obtained it in 2013, from UGA. After a two-year post-doc period at the Multimodal Human Understanding Group, at the University of Trento, he was appointed Research Scientist at Inria until 2024 (when he was appointed Research Director). Xavier is an active member of SIGMM, an IEEE Senior Member, and an ELLIS Fellow. He was the Coordinator of the terminated H2020 Project SPRING: Socially Pertinent Robots in Gerontological Healthcare and co-lead the "Audio-visual machine perception and interaction for companion robots" chair of the Multidisciplinary Institute of Artificial Intelligence. He has co-supervised more than 20 PhD students and published over 120 papers in leading venues including IEEE TPAMI, CVPR, NeurIPS, ECCV, ICASSP, and ACM Multimedia. Xavier’s research interests are at the crossroads of machine learning, computer vision, and audio processing for scene and behavior analysis and human-robot interaction.

Extended Bio

My full name is Xavier Alameda-Pineda, I was born in Barcelona, Catalunya, Spain. I lived for 24 years in Cornellà del Llobregat, a city really close to the wonderful Barcelona. So, I am strongly linked to that country. Wherever I live, wherever I travel, I take its culture with me, trying to mix it with the local one. Because I was born in Catalunya, I am bilingual since I was a little kid. In addition to Catalan and Spanish, I also speak English, French and Italian (from my time in Trento). French is my every-day language.

Undergrad & Masters at BarcelonaTech

I obtained a (5-year) degree in Mathematics from the Faculty of Mathematics and Statistics in 2008 and a (5-year) degree in Telecommunications Engineering from the School of Telecommunications in 2009 both from the Technical University of Catalonia — BarcelonaTech. I had a multidisciplinary education in Mathematics, Telecommunications Engineering and Computer Science. My first steps in research were in functional analysis and in signal processing. On one hand, and supervised by Prof. Xavier Cabré, I worked on non-linear partial differential equations in the Department of Applied Mathematics 1. On the other hand, under the supervision of Prof. Philippe Salembier, I explored the use of the Generalized Lifting transform for image compression in the Signal Theory and Communications Department.

Masters and PhD at Inria

Afterward, I went to Grenoble to follow the MoSIG masters program in Graphics, Vision and Robotics, offered at UFR IM2AG from University Joseph Fourier (Grenoble 1) and Grenoble INP – Ensimag (formerly ENSIMAG, INP Grenoble). I got my PhD in October 2013, at the Perception Team at Inria Grenoble Rhône-Alpes, under the supervision of Prof. Radu Horaud, within the framework of the HUMAVIPS project. I worked on AV speaker detection and localization, Geometric sound source localization, The RAVEL data set and AV robot gesture recognition.

Postdoctoral Positions

I spent one year collaborating with the CRISSP team of GIPSA-Lab (CNRS), and with Prof. Laurent Girin in particular, in the following topics: Adaptation of Gaussian mixture regressors for acoustic-articulatory inversion and Probabilistic methods for sound source separation and sound source modeling.

From 2014 to 2016, I was a post-doctoral fellow in the Multimodal Human Understanding Group, in University of Trento, with Prof. Nicu Sebe working on: The automatic analysis of social interactions with multimodal data (using the SALSA dataset), Weighted-data algorithms for clustering with applications to active speaker detection, originally-paced learning methods for sketch-based image retrieval and Human behavior analysis tasks based on vision (remote heart-rate, emotion recognition).

Hobbies — when I don’t research

I like many different things: most music styles, several sports, almost any existing food/drink that I have ever tasted and many different dances. I also enjoy Police/Crime books/TV series and movies. Music: classical, flamenco, jazzy and catalan rumba. Mountain: I like climbing/bouldering and hiking in the mountains near Grenoble, either in Chartreuse, in Vercors or in Belledonne. I also enjoyed skying near Trento. You can check out here many different outdoor experiences I’ve gone through. Food: when the weather is not good, I may stay home to cook some good food such as "paella", "tartiflette", "tortilla" or "pà amb tomàquet". Dancing: In the evening you will probably find me enjoying some dancing swing. I took dancing courses in Grenoble Swing for two years, and then keep on dancing and taught in Swing Dance Trento. It is a sportive and funny dance. You should give it a try! Currently, I am dancing and teaching swing dancing back in Grenoble Swing.