Overview
Prof Mikolajczyk's research activities are in computer vision and machine learning for visual recognition, retrieval and matching. His particular interests are in invariant feature extraction and visual signal representation which he applied to many computer vision tasks including image and video classification, object recognition and tracking, visual surveillance, large scale image retrieval, wide baseline matching, face recognition, 3D reconstruction, segmentation, etc.
MatchLab Research Group
RESEARCH HIGHLIGHTS
IPALM, Interactive Perception-Action-Learning for Modelling Objects
FACER2VM, Face Matching for Automatic Identity Retrieval, Recognition, Verification and Management
RESEARCH GRANTS
- EPSRC Grants
- 2019-22 EPSRC, Interactive Perception-Action Loop for Modelling Objects
- 2016-20 EPSRC, Face Matching for Automatic Identity Retrieval, Recognition, Verification and Management
- 2013-16 EU Chist-ERA, EPSRC, Tagging Visual data with semantic descriptors
- 2008-2013 EPSRC UK Case, British Broadcasting Corporation (BBC), - development of a system for image retrieval in large databases
- 2012-14 EPSRC UK, Proof of Concept for Animal health monitoring
- 2008-10 EPSRC UK, Recognition of object classes and scenes
- 2008-10 BAE Systems, Biometric Security System
- 2008-10 EU project, IDASH, The Investigator Dashboard
- 2009-14 EPSRC UK, Visual Media Research Platform
- 2006-09: European project, Vidi-Video, Interactive Semantic Video Search
PEOPLE
RESEARCH ASSISTANTS
- Dr Adrian Lopez Rodriguez
- Dr Yurun Tian, Research Associate
- Dr Fei Yan, Senior RA Machine Learning for Image Classification
- Dr Mark Barnard, post-doc Sport Action Recognition
- Dr Svetlana Barsky, post-doc Location identification in large forensic databases
- Dr Muhammad Tahir, post-doc, Linear and non-linear classifiers
PHD STUDENTS
- Tony Ng, Camera Localisation
- Michal Nazarczuk, Robotics Reasoning
- Mikolaj Jankowski, Wireless Edge Computing
- Roy Miles, Knowledge Distillation
- Dylan Auty, Monocular depth estimation
- Adrian Lopez Rodriguez, Domain adaptation
- Sara Iodice, Person re-identification
- Vassilis Balntas, Efficient image descriptors
- Aarushi Gaur, Automatic assessment of visual aesthetics
- Falk Schubert, Super-resolution and image enhancement based on priors and learning
- Peter Koniusz, Modelling object categories using image segmentations
- Muhamad Rana, Information fusion for image classification
- Zdenek Kalal, PhD student: Online training and tracking of objects
- Honping Cai, Optimal linear projections for dimensionality reduction of image descriptors
- Hirofumi Uemura, Learning motion-appearance models of action categories
VISITING PHD STUDENTS
- Edgar Riba, Universidad Autonoma de Barcelona, Image Descriptors
- Jiaolong Xu, Universidad Autonoma de Barcelona, Spain, Pedestrian detection for driving assistance
- Osman Akin, Hacettepe University, Turkey, Multiple objects tracking
- Alvaro Gonzalez Arroyo, University of Alcala, Spain, Text recognition
- David Geronimo, Universidad Autonoma de Barcelona, Spain Pedestrian detection for driving assistance
COLLABORATIONS
Academic Collaborations
- Institut de Robotica i Informatica Industrial, Spain, Image analysis
- Aalto University, Finland
- ENPC ParisTech, France
- Czech Technical University, Czech Republic
- University of Sheffield, UK, Language and vision processing
- Ecole Centrale de Lyon, France, Image descriptors
- Kyushu Institute of Technology, Japan, visiting scholar, co-supervision of a PhD student in the domain of human action recognition
- Universidad Autonoma de Barcelona, Spain, co-supervision of visiting PhD students in the domain of pedestrian detection and surveillance
- Lappeenranta University of Technology, Finland,
- Tampere University of Technology, Finland
- University of Leon, Spain, visiting scholar
- University of Amsterdam, Holland, Collaboration in international recognition competitions
- University of Oxford, UK, image features and matching benchmarks
- Technical University of Brno, Czech Republic, co-supervision of PhD, Master and B.Eng students in fast feature matching
- Czech Technical University of Prague, Czech Republic, co-supervision of a PhD student in the domain of computer vision
- Universidad de Alcala, Madrid, Spain, co-supervision of a PhD student in text detection
Industrial Collaborations
- BBC, UK - Video recognition for multimedia archives.
- Yazaki - tracking system.
- NDA, USA - Tracking Learning Detection for Multimedia applications.
- BAE Systems, UK - Trademark detection system.
- Metropolitan Police of London, UK - Video analysis technology and commercial solutions.
- SWARM Systems Ltd, UK - Object recognition system for UAVs.
- Zoological Society of London, Cambridge Consultants Ltd., Seven Technologies Ltd - Surveillance cameras for animal conservation.
- APHA, UK - Animal and Plant Health Agency.
- Marwell ZOO, Phenotyping large exotic species.
- Vistar Ltd. - Night vision system for marine use.
- Complex Systems Innovations Ltd, UK dense image registration.
- Evolution Robotics, USA - Evaluation of a target recognition system.
INVITED TALKS
Invited Talks
- Everything Has Been Done, It Is Our Job to Do It One Better in Image Matching and Localisation, ICPRAM, 2022
- Image features 20 years on, Camera Localization Workshop, CVPR, June 2020
- Recent progress in local descriptors, Retrieval Workshop, CVPR, Long Beach, June 2019
- Domain Adaptation and Few-Shot Learning, ACCV Workshop 2018
- Is image to image matching solved? High Visual Computing 2018
- International Symposium on Computer and Information Sciences, Krakow, Poland, 2016
- Vision and Language Integration, EMNLP Workshop, Lisbon, Portugal, 2015
- International Conference of Artificial Intelligence and Soft Comp., Poland, June 2015,
- XXXV Jornadas de Automatica, Valencia, Spain, Sept 2014,
- International Conference on Pattern Recognition, Tsukuba, Japan, Nov 2012,
- InstantWILD, a revolution in field conservation, Zoological Society of London, UK, April 2014,
- MOSES Computer Vision Workshop, Siegen, Germany, Oct 2011,
Seminars
- Online learning and tracking, University of Exeter UK, Feb 2013;
- Local features; Kyushu Institute of Technology, Kyushu, Japan, November 2012;
- Tracking Learning Detection, Imperial College, UK, Nov 2012;
- PN learning for tracking objects. INRIA Paris, France Nov 2010;
- Online object recognition, BBC London, UK, Oct 2010;
- Image classification and retrieval for video, BBC Kingswood, UK, Nov 2009;
- Multi-action recognition and localization, University of Southampton, UK, July 2009;
- Target recognition for UAV vehicles, EPFL, Switzerland, Feb 2009;
- Super-resolution and high dynamic range imaging, Cranfield University, UK, Feb 2009;
- Action recognition with a vocabulary forest, INRIA Rhone-Alpes, March 2008;
- Statistical methods for learning image descriptor projections, Universidad Autonoma de Barcelona, Spain, Feb 2008;
- Category representation for recognition and tracking; Imperial College, UK, May 2007;
- Object and action class recognition; University of Lappeenranta, Finland, April 2007;
- Linear discriminant projections of local descriptors for dimensionality reduction; Technical University of Prague, Czech Republic, April 2007
- Learning of object class models; University of Amsterdam, Holland, Feb 2007;
- Learning of decision trees for visual recognition; Kyushu Institute of Technology, Kyushu, Japan, August 2006
- Image representation for recognition and tracking of objects; British Machine Vision Association Workshop, London UK, June 2006;
- Image retrieval from large database; University of Oxford, May 2006;
- Human detection by a probabilistic assembly of parts; ETH, Switzerland, Aug 2004.
Tutorials
- BMVA Summer School 2h, Manchester, UK, July since 2010
- iV&L Net Training School, Malta, 2016
- Winter School on Surgical Imaging and Vision 2h, Imperial College, UK, Dec 2013
- VISUM Summer School 5h, Porto, Portugal, July 2013.
- Vision and Language Workshop in Sheffield, 2h, Nov 2012.
- Modern features: advances, applications, and software 3h, European Conference on Computer Vision, Florence, Italy, Nov. 2012
- AERFAISS Summer school 5h, Castallo de la Plana, Spain, June 2010
- Computer Vision and Machine Learning 24h, Kuala-Lumpur, Malaysia, June 2009
- Visual Object Recognition 3h, International Conference on Computer Vision, Rio de Janeiro, Brazil, Oct 2007
CONFERENCE/EDITORIAL SERVICES
General and Programme Chairing
- British Machine Vision Conference (BMVC), London, September 2017.
- Vision and Language workshop 2017
- ECCV Workshop Local Features: State of the Art, Open Problems and Performance Evaluation, Amsterdam, NL, October 2016.
- IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS), Krakow Poland, September 2013
- British Machine Vision Conference (BMVC), Guildford UK, September 2012.
- Workshop on Vision and Language, at Computer Vision and Pattern Recognition, Boston, USA, June 2015.
- Workshop on Feature Detectors and Descriptors: The State Of The Art and Beyond, at Computer Vision and Pattern Recognition, Miami, USA, June 2009.
- Workshop on Object Recognition, British Machine Vision Association, London, UK, March 2006.
- Workshop on Emerging Vision Applications, at World Multiconference on Systemics, Cybernetics and Informatics, ISAS-SCIs, Orlando, USA, July 2001.
Area Chair
- International Conference on Computer Vision, Santiago, Chile, 2015.
- Conference on Computer Vision and Pattern Recognition, Boston, USA, 2015.
- International Conference on Computer Vision, Barcelona, Spain, 2011.
- British Machine Vision Conference, since 2011.
Programme Committee Member
- International Conference on Computer Vision,
- European Conference on Computer Vision,
- IEEE Conference on Computer Vision and Pattern Recognition,
- British Machine Vision Conference,
- International Conference on Pattern Recognition
and other international conferences and workshops.
Journal Reviewer
- Guest Associate Editor of International Journal of Computer Vision,
- IEEE Transactions on Pattern Analysis and Machine Intelligence,
- IEEE Transactions on Image Processing,
- Journal of Machine Learning Research,
- Image and Vision Computing,
- Computers and Artificial Intelligence.
FUNDING ORGANISATION COMMITTEES
PHD APPLICATION PROCESS
The studentships are for three years starting typically Sept/Oct or sooner if possible. It covers University tuition fees (at EU/UK level) and provides an annual tax free stipend. For exceptional overseas candidates there is the opportunity to obtain additional funding to cover the increased fees. To apply for a PhD under my supervision
1. Read here
https://www.imperial.ac.uk/electrical-engineering/study/phd/
2. Create an account on College admission pages
3. Choose Postgraduate Research at the top,
find programme Electrical Engineering Research - Full Time
4. Start the application
Proposed Research Topic: Computer Vision
Proposed Research Group/Department/Centre: Electronic and Electrical Engineering/Intelligent Systems and Networks.
Proposed Research Supervisor: Krystian Mikolajczyk
Second Choice Programme Title: Electrical Engineering Research - Full Time,
Prepare your research statement in the area of Computer Vision and Machine Learning
Fill in the application and submit.
Email K.Mikolajczyk@imperial.... once this is done.