Research
Manuele Bicego's research is mainly focused on Pattern Recognition, namely on the study and development of automatic techniques and models able to extract information from real world data, typically in terms of classes or clusters. In the past he worked a lot on probabilistic models - like Hidden Markov Models, Mixtures, Topic Models - and on kernel machines - like Support Vector Machines. In these contexts he designed novel models/methodologies, like hybrid generative-discriminative methods, generative embeddings and kernels, novel classification schemes, model selection techniques and others. Currently he is very interested in problems related to unsupervised learning, such as clustering and biclustering. In this context, in recent years he started to investigate approaches based on Random Forests, flexible data description tools which exploitation for unsupervised learning is far from being as mature as in classification/regression.
At the same time, he is involved in application-driven research, where the novel approaches can be tested with challenging real world problems, as well as adapted and tailored in order to derive specific context dependent solutions. In the past he worked a lot on computer vision problems as object recognition from images, videosurveillance and video analysis, person recognition and authentication from images and videos (and, more in general, biometrics). Currently, he is working on seismic signals analysis, bioinformatics and medical data analysis. In this last context, in recent years he started to develop Pattern Recognition e Machine Learning solutions for specific medical domains, like Neuroscience and Neurology, in particular in the context of diseases like brain tumors, Alzeimher or Multiple Sclerosis. The main goal in this research is to devise highly interpretable solutions, being interpretability of methods and solutions the most stringent need in nowadays bioinformatics and medical data analysis.
Some present and past collaborations:
Institutions inside Italy: Univ. di Verona (Dip. di Informatica Dip. di Biotecnologie, Dip. di Scienze della Vita e della Riproduzione, Dip. di Patologia e Diagnostica, Dip. di Neuroscienze, Biomedicina e Movimento), Univ. di Trento (Centro Interdipartimentale Mente/Cervello - CIMEC), Istituto Italiano di Tecnologia - IIT (Dip. di Pattern analysis and Computer Vision; Dip. di Drug and Discovery; Dip. di Nanoscienze), Univ. di Sassari (Computer Vision Lab, Dip. di Scienze Economiche e Aziendali), CRS4 (Data Intensive Computing group), CNR Sassari (Istituto di Chimica Biomolecolatre -- ICB), Univ. di Venezia (Dip. di Informatica), Univ. di Messina (Dip. di Scienze Cognitive e della Formazione).Institutions outside Italy: Delft University of Technology, Delft, The Netherlands, Universidad Nacional de Colombia, Manizales, Colombia, Instituto Superior Tecnico, Lisbon, Portugal, Universidad de Alicante, Alicante, Spain, Radboud University, Nijmegen, The Netherlands, Technical University of Dortmund, Dortmund, Germany, Microsof Research, Seattle, US, University of Oxford, Oxford, UK, ETH, Zurich, Switzerland, School of Computer Science, University of Manchester, Manchester, UK, Queen Mary College, London, UK, Bogazici University, Istanbul, Turkey, University of Vigo, Vigo, Spain, University of A Coruna, A Coruna, Spain, CWI (Center for Mathematics and Informatics), Amsterdam, The Netherlands
Main visit exchanges:
[2022, Apr-Sep]: Visit (5 months)Telecommunications Engineering School, University of Vigo, Vigo (Spain). [2021, Aug-Sep]: Visitor (1 month)
Marco Loog, Pattern Recognition & Bioinformatics Group of the Delft University of Technology of Delft (The Netherlands). [2018, March - Sep]: Visit (6.5 months)
Department of Computer Science and Artificial Intelligence, University of Alicante, Alicante (Spain) [2018, Feb]: Visitor (1 week)
Marco Loog, Pattern Recognition & Bioinformatics Group of the Delft University of Technology of Delft (The Netherlands). [2018, Feb]: Visitor (1 week)
David Tax, Pattern Recognition & Bioinformatics Group of the Delft University of Technology of Delft (The Netherlands). [2017, Sept-Dec]: Visitor (2.5 months)
Mauricio Orozco-Alzate, Department of Informatics and Computing of National University of Colombia (UNAL), Manizales (Colombia). [2016, Sep]: Visit (1 month)
Pattern and Image Analysis Lab of the Instituto Superior Tecnico (IST), University of Lisbon, Lisbon (Portugal). [2015, Sep]: Visit (1 week)
Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2015, May]: Visitor (1 month)
Marco Loog, Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2013, Sep]: Visitor (1 month)
Bob Duin, Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2013, Jun]: Visitor (1 month)
Mauricio Orozco-Alzate, Department of Informatics and Computing of National University of Colombia (UNAL), Manizales (Colombia). [2011, Oct-Dec]: Visit (3 months)
Department of Informatics and Computing of National University of Colombia (UNAL), Manizales (Colombia). [2009, Nov-Dec]: Visit (2 weeks)
Pattern Recognition and Telecommunications Lab of the Instituto Superior Tecnico (IST), Lisbon (Portugal). [2008, May-Jun]: Visit (2 weeks)
Signal Processing and Communications Dept. of the Vigo University of Vigo (Spain). [2007, Jun]: Visitor (1 week)
Bob Duin, Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2007, Jun]: Visitor (1 week)
David Tax, Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2007, Oct-Nov]: Visitor (1.5 months)
Daniel Gonzalez Jimenez, Signal Processing and Communications Dept. of the Vigo University of Vigo (Spain). [2007, Jan]: Visit (1 week)
Perceptual Intelligence Laboratory of the Bogazici University of Istanbul (Turkey). [2006, Nov-Dec]: Visit (1.5 months)
Information and Communication Theory Group of the Delft University of Technology of Delft (The Netherlands). [2006, Oct]: Visit (2 weeks)
Center for Mathematics and Computer Science (CWI) of Amsterdam (The Netherlands). [2005, Aug]: Visit (1 month)
Ecole Nationale Superieure des Telecommunications of Paris (France) to participate to the first BIOSECURE Residential Workshop, organized by BIOSECURE Network of Excellence. [2001, Sep-Dec]: Visit (4 months)
Pattern Recognition and Telecommunications Lab of the Istituto Superior Tecnico (IST), Lisbon (Portugal).
Scientific prizes and awards:
ICIAP 2023: Honorable Mention Best Paper AwardDescription: Award assigned to the 2nd best paper at the conference ICIAP 2023.
Details: Award received for the paper M. Bicego, M. Vazquez-Enriquez, J.L. Alba-Castro: "Active Class Selection for Dataset Acquisition in Sign Language Recognition", Proc. Int. Conf. on Image Analysis and Processing (ICIAP2023), (2023) SAC 2018: Best poster award
Description: Award assigned to the best 4 posters presented at the SAC 2018 conference.
Details: Award received for the paper A. Castellini, G.A. Beltrame, M. Bicego, J.J. Blum, M. Denitto, A. Farinelli: "Unsupervised Activity Recognition for Autonomous Water Drones", Proc. ACM Symposium on Applied Computing (SAC2018), pp. 840-842, (2018) CVPR 2014 Workshop on Biometrics: Highest Impact Award
Description: Award assigned to the paper published in a past edition of the CVPR Biometrics Workshop that has generated highest impact on the research community.
Details: Award received for the paper M. Bicego, A.Lagorio, E. Grosso, M. Tistarelli: On the use of SIFT features for face authentication, Proc. of IEEE CVPR Workshop on Biometrics (2006) ICPRAM 2012: Best student paper award (Area: Theory and Methods)
Description: Award assigned to the best paper of the ICPRAM 2012 conference, in the area of Theory and Methods, in which the first author is a PhD or MS student.
Details: Award received for the paper A. Carli, M. Figueiredo, M. Bicego, V. Muri- no: Generative Embeddings Based on Rician Mixtures: Application to Kernel-Based Dis- criminative Classification of Magnetic Resonance Images, Proc. of Int. Conf. on Pattern Recognition Applications and Methods (ICPRAM2012), (2012) ICIAP 2007: Best poster presentation award
Description: Award assigned to the best paper presented as a poster at the ICIAP 2007 conference.
Details: Award received for the paper M. Bicego, M. Cristani, V. Murino: Sparseness Achievement in Hidden Markov Models, Proc. Of IEEE Int. Conf. on Image Analysis and Processing (ICIAP 2007), pp. 67-72, (2007) University of Sassari 2007 Research award (Premio produttivita' scientifica 2007)
Description: Award assigned to the best 20 scientists of the whole University of Sassari (selected among all Full Professors, Associated Professors, Assistant Professors, Post Doc and PhD students of scientific areas).
Details: Award received for the years 2005-2006.