About

Xavier Alameda-Pineda is a researcher specializing in multimodal perception, human-robot interaction, and probabilistic machine learning, with a particular focus on audio-visual fusion for understanding human behavior in complex environments. His work has made significant contributions to the challenge of tracking, detecting, and localizing people using synchronized auditory and visual data — a problem central to building socially aware robotic systems. Among his most influential contributions are variational Bayesian frameworks for multi-person tracking in cluttered scenes, which together have garnered over 100 citations and represent a rigorous probabilistic approach to handling occlusions, appearance changes, and variable numbers of subjects. His early work on active-speaker detection using stereoscopic cameras and microphone arrays embedded in a robotic head (39 citations) demonstrated the practical power of sensory fusion in real-world robotics. The RAVEL corpus and his sound classification benchmarks laid important groundwork for training and evaluating domestic robots in realistic acoustic conditions. Across his career, Alameda-Pineda has consistently championed the complementarity of audio and visual modalities, showing that integrating both streams yields systems far more robust than either alone. With over 300 cumulative citations, his research has shaped modern approaches to perceptual intelligence in humanoid and companion robots.

Research Focus

Key Achievements

10
H-Index
18
Papers
362
Total Citations
20
Avg Citations/Paper
🏆 Most Cited Paper
Tracking Multiple Persons Based on a Variational Bayesian Model
68 citations · 2016
📈 Most Prolific Year: 2012 (4 Papers)
🤝 Key Collaborators: 67
🏛 Institutions: University of Trento, Institut national de recherche en sciences et technologies du numérique, Centre Inria de l'Université Grenoble Alpes, Université Grenoble Alpes, Institut polytechnique de Grenoble, Directorate-General for Interpretation

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 16 days ago