Dr. Li-Qun Xu

Future Content -

Research Theme Leader - Visual Computing and Multimedia Understanding

Li-Qun joined BT Laboratories in 1996 as a Senior Intelligent System Researcher and was promoted as Principal Researcher in Image and Visual Computing in 2001. He has been with BT's Research organisation while taking various leading research and project management roles in Perception and Coding Lab, Multimedia Applications, Broadband Applications Research Centres. His recent research interests and projects include broadly multimedia content analysis and indexing, robust object segmentation and tracking for intelligent visual surveillance, people behaviour and event analysis, motion analysis and segmentation, 3-D vision techniques and image-based rendering for collaborative working environment, among others.

Li-Qun holds BSc (1982), MSc(1985) and PhD^? (1988) degrees all in Electronic and Information Engineering. He worked for the Chinese Academy of Sciences in Beijing as a Research Assistant Professor in late 1980s prior to winning a Royal Society Research Fellowship and moving to the UK in early 1990. Between 1990 and 1996, he was working in Universities (UEA, Sussex, KCL, Leeds and Abertay Dundee) initially as member of Research Staff and lately as faculty.

He is currently an industry advisor to the on-going multi-million Euro EU FP6 Network of Excellence project KSPACE; an Executive Team member of IET Imaging and Vision Professional Network (I&V PN); an independent expert conducting research proposals evaluation and projects review for EU Commission Services (2002 - ) in ICT area of "Networked audio-visual system and home platform", "Semantic-based knowledge and content systems," and Security area of Visual-based surveillance as well as an EPSRC Peer Review College member (2006-2009). He is a Steering Committee member of conference series of SAMT and AVSS and serves regularly as Organisation and Programme Committee member of a number of international conferences and workshops. He is an Industry Advisory Board member of Department of Electronic Engineering, Queen Mary University of London (2007 - ) and has been on PhD^? thesis exam committees for TU-Delft (The Netherlands), UPC (Spain) and UCL (Belgium), and is currently an industry supervisor of 4 PhD^? studnets.

Projects

Network-centric smart systems and analytics for videos (NC-S2AV)
Visual scene analysis - Underground platform congestion level estimation
CCTV video enhancement through application of video super-resolution
Audio-visual semantic discovery in a multi-sensor environment
Modelling of crowd dynamics for ambient intelligence
Content management for user generated videos

Past Projects

Understanding Visual Content - Intelligent Video Surveillance
EU FP5 IST BUSMAN project
EU IST FP5 NoE SCHEMA
Understanding Visual Content - Multimedia Analysis and Digital Library

Recent Awards and Nominations

Finalist of 2008 BT Innovation Awards - 'Sir Alan Rudge Award for Future Innovation' with a patent entitled "Video Abstraction"
Finalist of 2007 BT Innovation Awards - 'The Gordon Radley Award for Best Author of Innovation', the entry paper is entitled "Affective video content representation and modeling"
Winner of 2006 BT Innovation Awards - 'Best Author for Innovation' Award with a paper entitled "Segmentation and tracking of multiple moving objects for intelligent video analysis"
Finalist of 2006 BT Innovation Awards - 'Best Pan-BT Future Growth Innovation' with a patent entitled "Object detection in images"

Principal fields of interest

Computer vision and applications (e.g. intelligent visual surveillance, object segmentation and tracking, motion estimation, 3D vision, image-based rendering)
Multimedia signal processing, indexing and retrieval (e.g. affective video content modelling, automated video summarisation and scene segmentation, semantics extraction, browsing and search in multimedia digital libraries, etc)
Multimodal interface and ambient intelligence (e.g. face detection / tracking, gaze analysis, expression/mode recognition)
Artificial neural networks, machine learning, data mining, pattern recognition
Networked audio-visual systems

He has 20 European and international patents granted or pending in these technical areas.

Selected recent publications

Li-Qun has authored and co-authored numerous research papers in professional journals and conference publications.

Li-Qun Xu and A, Anjulan, Crowd behaviours analysis in dynamic visual scenes of complex environment, IEEE ICIP Workshop on Multimedia Information Retrieval: New Trends and Challenges, San Diego, CA, October 2008.

B. Zhan, N.D. Monekosso, P. Remagnino, S.A. Velastin, L Xu, Crowd analysis - a survey, Machine Vision and Applications, Springer Berlin / Heidelberg, April 2008.

I. Tziakos, A. Cavallaro and Li-Qun Xu, Video events segmentation and visualisation in non-linear subspace, Pattern Recognition Letters, Elsevier Journal, 2008.

Li-Qun Xu, Issues in video analytics and surveillance systems: research / prototyping vs. applications / user requirements, in Proc. of IEEE Intl. Conference on Advanced Video and Signal-based Surveillance, September 2007, London, UK.

Li-Qun Xu and B. Luo, Appearance-based video clustering in 2D locality preserving projection subspace, in Proc. of ACM International Conference on Image and Video Retrieval, Amsterdam, The Netherlands, July 2007.

Li-Qun Xu and Pere Puig, Multi-level attention model for tracking and segmentation of humans under complex occlusion, in BT Technology Journal, April 2006.

Bangjun Lei and Li-Qun Xu, Real-time outdoor video surveillance with robust foreground extraction and object tracking via multi-state transition management, in Elsevier Publisher Journal, Pattern Recognition Letters, 2006.

Jerome Meessen, Li-Qun Xu, Benoit Macq, Content browsing and semantic context viewing through JPEG-2000 based scalable video summary, in IEE Journal: IEE Proceedings on Vision, Image & Signal Processing, Special Issue on Integration of Knowledge, Semantics and Digital Media Technology, April 2006.

Li-Qun Xu, Jose-Luis Landabaso, and Bangjun Lei, Detection and tracking of multiple moving objects for advanced visual surveillance, Chapter 15 in Intelligent Spaces: The Application of Pervasive ICT (eds. Alan Steventon and Steve Wright), Springer-Verlag London Limited, 2006.

Alan Hanjalic and Li-Qun Xu, Affective video content representation and modeling, in IEEE Transactions on Multimedia, Vol. 7, No. 1, February 2005.

Sergio Benini, Li-Qun Xu, Riccardo Leonardi, Using lateral ranking for motion-based video shot retrieval and dynamic content characterisation, in Proc. of 4th International Workshop on Content-Based Multimedia Indexing (CBMI'05), Riga, Latvia, 21-23 June, 2005.

Lectures & seminars

Panel discussion moderator at IEEE Intl. Conference on Advanced Video and Signal-based Surveillance (AVSS’07), London. Panel members include leading researchers from IBM, GE, Mitsubishi, Stanford University, HOSDB and two specialists’ video analytics firms, and over 150 international participants are present (05/09/07)
Invited seminar, E-Learning Lab, Shanghai Jiaotong University, China (29/03/07)
Invited speaker, Multimedia Knowledge Management Workshop: Industry meets academia, organised by Knowledge Media Institute, Open University. Over 70 participants from UK and European Universities and media technology industry (08/01/07)
Invited speaker, research seminar, Department of Electronics, Univ. of Kent (04/05/06)
Invited speaker, research seminar, Department of Computing, Univ. of Surrey (22/02/06)
Invited speaker, London Knowledge Management Seminar series, Department of Computing, Imperial College London(9/11/05).

General interests

Swimming, badminton, outdoor walking and cycling

Adastral Park, Martlesham Heath, Ipswich, Suffolk. IP5 3RE UK