Engineering, Computing & the Environment

Jean-Christophe Nebel

Machine Learning & Computer Vision Research

Current projects

Deep Fake Detection

Analysis of the Behaviour of Honey Bees from Audio Signals

Genomics-inspired Computer Vision

Past projects

Nonlinear dimensionality reduction methods applied to action recognition

Human pose recovery and tracking

MEDUSA - 3D posture recovery from video sequence

Michelangelo - 3D dynamic full body scanner development

V-Man - The Virtual Man

3D Thermography - 3D Thermography in the diagnostic of equine lameness

Current research students

Nour Eldin Alaa Badr, PhD student, since 2023
Deep Fake Detection for Human Faces

Supervision Team: Dr Xing Liang (FS), Prof. Jean-Christophe Nebel, Dr Greenhill

Peter Edwards, PhD student, since 2023
A Deep Learning Methodology for the Automatic Detection of Deepfake Media

Supervision Team: Dr Xing Liang (FS), Prof. Jean-Christophe Nebel, Dr Greenhill

Stenford Ruvinga, PhD student, since 2019
Bees as Environmental Sensors - The Application of Signal Processing and Machine Learning to Model, Quantify and Predict the Effects of Environmental Stressors on the Health and Well-Being of Bees

Supervision Team: Dr Gordon Hunter (FS), Dr Olga Duran, Prof. Jean-Christophe Nebel and Dr Rosa Busquets

Completed research students

Nati Ofir, PhD (2021)
Classic versus deep approaches to address computer vision challenges: A study of the problems of faint edge detection and multispectral image registration

Supervision Team: Prof. Jean-Christophe Nebel (FS)

Next position: Algorithm Developer at Applied Materials

Ioannis Kazantzidis, PhD (2019)
Genomics-inspired Computer Vision

Supervision Team: Dr Jean-Christophe Nebel (FS), Dr Francisco Florez-Revuelta, Dr Natasha Hill

Next position: Co-Founder & Research Lead at Yepic AI

Spyridon Bakas, PhD (2014)
Computer-Aided Localisation, Segmentation and Quantification of Focal Liver Lesions in Contrast Enhanced Ultrasound

Supervision Team: Dr Dimitrios Makris (FS), Dr Gordon Hunter, Dr Jean-Christophe Nebel

Next position: Postdoctoral Researcher at the University of Pennsylvania, USA

Alexandros Moutzouris, PhD (2013)
Pose Recovery from Multiple Cameras in Complex Scenarios

Supervision Team: Dr Dimitrios Makris (FS), Dr Jean-Christophe Nebel

Michal Lewandowski, PhD (2011)
Advanced dimensionality reduction methods applied to action recognition [PDF]

Supervision Team: Dr Jean-Christophe Nebel (FS), Dr Jarek Francik, Dr Dimitrios Makris, Dr James Orwell

Next position: Postdoctoral Researcher at Kingston University

Former researchers and visitors

Paul Kuo, Research assistant, 2006-2009
Posture recognition from video sequence

Sup. Team: J.-C. Nebel, D. Makris

Amin Dadgar, MSc student, 2007-2009
Pose recovery in context specific scenarios
Sup. Team: D. Makris (FS), J.-C. Nebel, T. Ellis

Jesus Martinez del Rincon, Vis. PhD stud., 07-08
Robust multi-target tracking (Comp. Vision Lab, Zaragoza U., Spain)
KU Sup. Team: J.-C. Nebel, D. Makris

Thibault Ammar, Visiting Master student, 2008
3D posture recovery (ISTASE, France)
KU Sup. Team: J.-C. Nebel, P. Kuo

Romain Dieny, Visiting Master student, 2010
Stereo matching (ENSPS, France)
KU Sup. Team: J.-C. Nebel, J. Martinez del Rincon

Francisco Martinez, Postgraduate researcher, 2010-2011
Action recognition
Sup. Team: J.-C Nebel, S. Velastin

Jerome Thevenon, Postgraduate researcher, 2010-2011
Action recognition
Sup. Team: J.-C. Nebel, S. Velastin

Maria Jose Santofimia Romero, Visiting Lecturer, 2011-2012
Action recognition (University of Castilla-La Mancha, Spain)
Sup. Team: J.-C Nebel, J. Martinez del Rincon

Jesus Martinez del Rincon, Research Fellow, 2009-12
3D Articulated Tracking
Sup. Team: D. Makris, J.-C. Nebel

Ana Carolina dos-Santos-Paulino, Visiting Master student, 2013
Evolutionary algorithm for dense pixel matching (Telecom Physique Strasbourg)
Sup. Team: F. Florez-Revuelta, J.-C. Nebel

Mickael Dequidt, Visiting Master student, 2014
Foreground extraction (ENSICAEN, France)
Sup. Team: J.-C. Nebel

Hoa-Cuc Nguyen-Thi , Research student, 2014-2018
Recognition of activities of daily living with wearable vision systems
Sup. Team: F. Florez-Revuelta, J.-C. Nebel

Videos

Single Action Manifold for Walking and Running actions

Generated using Generalised Laplacian Eigenmaps (GLE) (see publication for details)

View and Style-Independent Action Manifold ('kick' action)

Generated using Temporal Extension of Laplacian Eigenmaps (TLE) (see publication for details)
Training data: action sequences from IXMAS dataset (Computer Vision & Image Understanding. 104(2-3), 2006)

2D Manifold of a walking cycle

Generated using Temporal Extension of Laplacian Eigenmaps (TLE) (see publication for details)
Training data: mocap sequences from HumanEva dataset (Int. J. Computer Vision, 87(1-2), 2010)

Human Interaction Recognition

Using Bag Of Words approach (BoW)
Video data: interaction sequences from the UT-Interaction dataset (ICCV2009)

Activity independent 3D pose recovery

Exploiting human bipedal motion constraints (see publication for details)
Video data: walking sequences from HumanEva dataset (Int. J. Computer Vision, 87(1-2), 2010)

Activity independent 2D pose tracking

Using particle filters constrained by human biomechanics (see publication for details)
Video data: walking and balancing sequences from HumanEva dataset (Int. J. Computer Vision, 87(1-2), 2010)

Available datasets

Distorded image Data used to evaluate quality of dense pixel matching algorithm on distorted images [PDF].

Media

Los ordenadores son vigilantes novatos; la genomica les enseña a ser mejores, by Jean-Christophe Nebel & Francisco Florez Revuelta
The Conversation Spain, October 30, 2018 23.35 CET (over 1,000 readers)

"La videovigilancia automatizada todavia no es fiable, pero las tecnicas de analisis genetico pueden ayudar". Our bioinformatics research is featured in this article.

Reconnaitre des visages dans une video en s’inspirant de l’analyse de l’ADN, by Jean-Christophe Nebel
The Conversation France, November 6, 2017 10.39pm GMT (over 2,500 readers)

"La videoprotection est censee assurer la securite du public. Mais comment traiter des sequences de plus en plus nombreuses ? En considérant les images animees comme des mutations genetiques…" Our research on 'Genomics-inspired Computer Vision' is featured in this article.

DNA techniques could transform facial recognition technology, by Jean-Christophe Nebel
The Conversation UK, October 20, 2017 2.36pm BST (over 24,000 readers)

"Treating video like a mutating gene could improve surveillance software." Our research on 'Genomics-inspired Computer Vision' is featured in this article.
This article was republished in Scientific American

Decoding the Language of Human Movement, by Chris Edwards
Communications of the ACM, Vol. 57 No. 12, Pages 56-67, December 2014

"Computers that recognize what is happening in moving images can help defend against crime, and revolutionize rehabilitation."
Our research on 'Episodic Reasoning for Vision-Based Human Action Recognition' is featured in this article.

Des cameras intelligentes pour surveiller le quartier des Paquis a Geneve
RTS INFO, Friday, 9 August 2013

Combattre la criminalite avec des cameras capables de detecter des comportements suspects. Geneve n'exclut pas de recourir a la videosurveillance dite "intelligente" dans son projet pilote pour les Paquis. Les autorites genevoises n'ont pas pris de decision finale. Le Grand Conseil se penchera sur la question surement cet automne. Mais quelles sont les capacites et limites de la videosurveillance intelligente? L'enquete de Jordan Davis.

New gun spotting role for CCTV
BBC News 24, Tuesday, 20 March 2007

Experts are investigating whether CCTV cameras could be used to see if a person is carrying a weapon.

Impact

Economic benefits from sales of people-tracking and crowd-monitoring technology
Research Excellence Framework 2014

"Research at Kingston University into methods for tracking pedestrians and monitoring crowds using computer vision techniques has been translated into commercial products by Ipsotek Ltd and BAe Systems, resulting in economic benefits to these companies from sales of these products. These products have been sold to high-profile customers including the London Eye, the O2 Arena and the Australian Government, providing significant commercial benefits, employment and growth for both companies, as well as providing an economic impact for these customers."

Underpinning research includes: Tracking Human Position and Lower Body Parts Using Kalman and Particle Filters Constrained by Human Biomechanics, J. Martinez del Rincon, D. Makris, C. Orrite-Urunuela and J.-C. Nebel, IEEE Transactions on Systems Man and Cybernetics - Part B', 40(4), 2010 [PDF]

Publications

Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

j.nebel@kingston.ac.uk