social linkedin box blue 32
social facebook box blue 32
social twitter box blue 32
social facebook box blue 32
iit-logo-v3-mainsite
People ■ Sean Ryan Fanello
Sean Ryan Fanello's foto

Sean Ryan Fanello

PhD Student

Fellow-PhD

iCub Facility


Address Microsoft Research 21 Station Rd
City Cambridge ZIP Code CB1 2FB
Country United Kingdom
Mobile Phone +44 7456160730
IM Address seanryanpro (Skype)

Bio

Sean Ryan Fanello was born in Cinquefrondi (RC) in 1987. He completed his Bachelor's Degree in Computer Engineering with the score of 110/110 cum laude at Sapienza, University of Rome, in September 2008. In 2010 he graduated (cum laude) in Computer Engineering with a specialisation in Artificial Intelligence. He worked on his Master Thesis at Imperial College of London, with the Intelligent Systems and Networks group under the supervision of Prof. Yiannis Demiris. He developed a Real-Time Arm-Hand Gesture Recognition System for Humanoid Robots, winning the Best Paper Award at ISVC 2010. From January 2011 to December 2013 he carried out PhD studies on Robotics, Cognition and Interaction Technologies at Italian Institute of Technology. He was in charge of developing theory, algorithms and applications for the visual perception system of the iCub humanoid robot. In particular he worked on image representations for object and action recognition, motion perception, 3D estimation and eye-hand calibration.

From July 2013 to November 2013 he was a Research Intern at Microsoft Research in Cambridge, UK. He is currently a Research Consultant at Microsoft Research.

Sean Ryan Fanello was born in Cinquefrondi (RC) in 1987. He completed his Bachelor's Degree in Computer Engineering with the score of 110/110 cum laude at Sapienza, University of Rome, in September 2008. In 2010 he graduated (cum laude) in Computer Engineering with a specialisation in Artificial Intelligence. He worked on his Master Thesis at Imperial College of London, with the Intelligent Systems and Networks group under the supervision of Prof. Yiannis Demiris. He developed a Real-Time Arm-Hand ...

●•. Read More

Projects

Object Categorization with iCub

The goal of the project was to develop new techniques for learning and recognition of object categories. We started from current state of the art visual recognition pipelines based on sparse representations, then we developed a new hierarchical image representation. The theoretical concepts of sparsity and locality have been used in order to increase the discrimantive power of the image descriptors, still maintaning real-time performances. Results are shown here:

http://www.youtube.com/watch?v=ZIpVrD6e-kA

 

iCub World Data-set

The iCubWorld is an object categorization data-set of 10 categories and 40 objects. The goal is to teach iCub to recognize objects from one-shot video sequences in Human-Robot-Interaction (HRI) scenarios. Despite the size it is very challenging and state of the art computer vision algorithms achieves very poor results. This is due to the absence of helpful context as in common object recognition benchmarks.  

The data-set can be downloaded from: http://www.iit.it/it/projects/data-sets.html

An example of the acquisition process can be found at: http://www.youtube.com/watch?v=vhPLUNg9r5k

 

All Gestures You Can: A Memory Game

“All Gestures You Can”  is a gesture recognition game aiming at improving memory skills, visual association and concentration. Our game takes inspiration from the classic “Simon” game; nevertheless, since the original version has been often defined as “visually boring”, we developed a revisited version, based on gesture recognition, which involves a “less boring” opponent: the iCub. The game won the second prize at Kinect Demonstration Competition - CVPR 2012, see the official website. Also a video of the game can be found at http://www.youtube.com/watch?v=U_JLoe_fT3I

 

Object Categorization with iCub

The goal of the project was to develop new techniques for learning and recognition of object categories. We started from current state of the art visual recognition pipelines based on sparse representations, then we developed a new hierarchical image representation. The theoretical concepts of sparsity and locality have been used in order to increase the discrimantive power of the image descriptors, still maintaning real-time performances. Results are shown here:

●•. Read More

Selected Publications

International Journals:
 
S.R. Fanello, I. Gori, G. Metta, F. Odone
International Journal of Machine Learning Research (JMLR), 2013
 
International Conferences:
 
S.R. Fanello, N. Noceti, G. Metta, F. Odone
International Conference on Computer Vision Theory and Application (VISAPP) 2014 
 
C. Ciliberto, S.R. Fanello, M. Santoro, L. Natale, G. Metta,L. Rosasco
IEEE International Conference on Intelligent Robots and Systems (IROS) 2013
 
S.R. Fanello, C. Ciliberto, M. Santoro, L. Natale, G. Metta, L. Rosasco, F. Odone
IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPR) 2013
 
S.R. Fanello, I. Gori, G. Metta, F. Odone
Iberian Conference on Pattern Recognition and Image Analysis (IbPRIA) 2013
 
S.R. Fanello, C. Ciliberto, L. Natale, G. Metta
IEEE International Conference on Robotics and Automation (ICRA) 2013
 
S.R. Fanello, N. Noceti, G. Metta, F. Odone
International Conference on Computer Vision Theory and Application (VISAPP) 2013
 
I. Gori, S.R. Fanello, F. Odone, G. Metta
IEEE Workshop on Robot Vision (WoRV) 2013
 
I. Gori, S.R. Fanello, G. Metta, F. Odone
IEEE-RAS International Conference on Humanoid Robots 2012
 
C. Ciliberto, S.R. Fanello, L. Natale, G. Metta
IEEE International Conference on Intelligent Robots and Systems (IROS) 2012
 
S.R. Fanello, l. Gori, F. Pirri
ISVC Conference on Computer Vision, Computer Graphics, Virtual Reality and Visualization 2012
 
A. Carrano, V. D’Angelo, S.R. Fanello, I. Gori, F. Pirri, A. Rudi
IASTED Conference on Signal Processing, Pattern Recognition and Applications 2010
International Journals:
 
S.R. Fanello, I. Gori, G. Metta, F. Odone
International Journal of Machine Learning Research (JMLR), 2013
 
International Conferences:
 
S.R. Fanello, N. Noceti, G. Metta, F. Odone
International Conference on Computer Vision Theory and Application (VISAPP) 2014 
 
C. Ciliberto, S.R. Fanello, M. Santoro, L. Natale, G. Metta,L. Rosasco

Awards

Special Mention at ibPRIA 2013
Honorable Mention for the oral presentation of the paper "One-Shot Learning for Real-Time Action Recognition".

Best Paper Award at WoRV 2013
Prize for the paper "A Compositional Approach for 3D Arm-Hand Action Recognition", WoRV Clearwater FL, 2013.

2nd place at Microsoft Kinect Demonstration Competition - CVPR 2012
Organizer: ChaLearn
Website: http://gesture.chalearn.org/dissemination/cvpr2012/demonstration-competition
Video: http://www.youtube.com/watch?v=U_JLoe_fT3I&

Best Paper Award at ISVC 2010
Prize for the paper "Arm-Hand behaviours modelling: from attention to imitation", ISVC, Las Vegas 2010.

Student Excellence Program
Sapienza University of Rome
This Scholarship is a special award for outstanding students in terms of average grade, exams passed and age. Furthemore, chosen students are entitled to attend special seminar courses and to participate in laboratory activities.



 

Special Mention at ibPRIA 2013
Honorable Mention for the oral presentation of the paper "One-Shot Learning for Real-Time Action Recognition".

Best Paper Award at WoRV 2013
Prize for the paper "A Compositional Approach for 3D Arm-Hand Action Recognition", WoRV Clearwater FL, 2013.

2nd place at Microsoft Kinect Demonstration Competition - CVPR 2012
Organizer: ChaLearn
Website: http://gesture.chalearn.org/dissemination/cvpr2012/demonstration-competition
Video: http://www.youtube.com/watch?v=U_JLoe_fT3I&

●•. Read More

Former colleagues