ACM SIGMM Rising Star Award 2018

The 2018 winner of the prestigious ACM Special Interest Group on Multimedia (SIGMM) Rising Star Award is Dr. Xavier Alameda-Pineda. The award is given in recognition of his contribution to multimodal social behavior understanding. 

 

Xavier obtained a Masters degree in Mathematics (2008), in Telecommunications (2009) and in Computer Science (2010), and a PhD in Mathematics and Computer Science (2013). His scientific career started very early, and he was awarded with several undergraduate scholarships as well a PhD fellowship from the French Ministry of Research and High Education. After his PhD under the supervision or Dr. Radu Horaud, he moved to Italy for a two-year post-doc period at University of Trento in the Multimodal Human Understanding Group, lead by Prof. Nicu Sebe. Since 2016, Xavier is a research scientist at Inria Grenoble Rhône-Alpes, in the Perception Team. 

 

Xavier’s main scientific interests are in computer vision, audio and signal processing, machine learning; and specifically how to combine these three disciplines for behavior analysis and synthesis and for smart interactive systems. Co-author of roughly 50 scientific publications, including a book on Multimodal Behavior Analysis in the Wild and a US patent for heart rate estimation from face videos under realistic conditions, Xavier’s publications received several awards, namely: the Outstanding Paper Award at ACM ICMI’11, the Best Paper Award at ACM MM’15, the Best Student Paper Award at IEEE WASPAA’15, the Best Scientific Paper Award on Image, Speech, Signal and Video Processing at IAPR ICPR’16, and was a Novel Technology Paper Award Finalist at IROS’17. 

 

From a methodological perspective, Xavier made a consistent contribution in three different directions. (i) The use of probabilistic models for auditory, visual and audio-visual multi-speaker scene analysis (tracking, sound separation, diarization, etc) and their combination with deep neural architecures, which lead to five top-journal papers and several conference papers in the past three years, and gathered 200+ citations. (ii) The use of low-rank models for missing data problems, that lead to several matrix completion algorithms for a plethora of multimedia tasks (3 top-level conferences with 150+ citations in the past three years). (iii) The collection, processing and publications of multi-modal datasets for scene analysis (2 journal papers with 80+ citations). 

 

The merit of these scientific contributions is shared with the PhD and MSc students that worked very hard under the co-guidance of Xavier together with Dr. Radu Horaud, Prof. Nicu Sebe or Prof. Elisa Ricci (Fondazione Bruno Kessler and University of Trento). Indeed, ,in the past five years, Xavier co-supervised roughly 10 MSc and 10 PhD students. Most of them are now pursuing their research and egineering careers either in large industrial research groups or in top academic computer vision and multimedia scientific groups. 

 

Xavier is also highly involved in the multimedia and computer vision communities. He systematically reviews for the main conferences and journals of the community (ACM MM, CVPR, ICCV, ECCV, ACM TOMM, IEEE TMM, IEEE TPAMI and IJCV). He was area chair of ICCV’17, and the co-chaired different scientific events including: two workshops at ACM MM (‘17, ‘18),  a special session at ACM ICMR’17, a workshop at IEEE CVPR’18 and a workshop at ECCV’18. He was also Guest Editor of two special issues at ACM TOMM and IJCV. For the past two years, Xavier contributes, jointly with Dr. Miriam Redi, to the Community Opinion Column of the ACM SIGMM Records. 

 

ACM is the professional society of computer scientists, and SIGMM is the special interest group on multimedia.