Acronym LTT
Funding Reference FCT - POSI/ CPS / 37844 / 2001
Dates 2002|2005
Summary

This project aims to develop methods for long term tracking of multiple objects in video sequences. Multiple object tracking has received the attention of the image processing community in the last 5 years, fostered by surveillance applications and by Model Based Video Coding (MPEG).
The first works addressed short-term tracking and recognition of activities. More recent works have tried to address long term tracking of moving objects. This is a more difficult problem since it involves the ability to disambiguate the trajectories of the objects after they were grouped and occluded for some time.
This project aims to address this problem. We wish to detect moving regions in video sequences and to develop algorithms to label each region in a consistent way along the whole video sequence. An additional difficulty concerns the presence of merged regions which can not be identified by a single label. Probabilistic models, namely probabilistic networks, will be adopted to perform this task and to propagate probable labelling scenarios. The tracking algorithms will be applied in the context of urban surveillance.

Research Groups Signal and Image Processing Group (SIPG)
Project Partners Polytechnic Institute of Lisbon
ISR/IST Responsible
Jorge S. Marques
[1] Gilles Celeux, Jorge S. Marques, Jacinto Nascimento, Learning Switching Dynamic Models for Objects Tracking, Pattern Recognition, Vol. 37, no. 9, 1835-1840, 2004
[2] Jacinto Nascimento, Arnaldo J. Abrantes, Jorge S. Marques, Using middle level features for robust shape tracking, Pattern Recognition Letters, Vol. 24, 295-307, 2003
[3] Jacinto Nascimento, Jorge S. Marques, Robust Shape Tracking in the Presence of Cluttered Background, IEEE Transactions on Multimedia, Vol. 6, no. 6, 2004
[4] Jorge S. Marques, Jacinto Nascimento, An Adaptive Potential for Robust Shape Estimation, Image and Vision Computing Journal, 21,1107-1116, 2003
[5] J. G. Silva, Jorge S. Marques, J. M. Lemos, Nonlinear Dimension Reduction with Tangent Bundle Approximation, Proc. of ICASSP2005 - IEEE International Conference on Acoustics, Speech and Signal Processing, Vol. 4, pp. 85-88, Philadelphia, USA, 2005
[6] Jacinto Nascimento, Mario Figueiredo, Jorge S. Marques, Recognition of Human Activities Using Space Dependent Switched Dynamical Models, Proc. of ICIP2005 - IEEE International Conference on Image Processing, pp. 852-855, Genoa, Italy, 2005
[7] Pedro Jorge, Arnaldo Abrantes, Jorge S. Marques, Tracking with Bayesian Networks. Extension to Arbitrary Topologies, Proc. of ICIP2005 - IEEE International Conference on Image Processing, pp. 402-405, Genoa, Italy, 2005
[8] Jacinto Nascimento, Mario Figueiredo, Jorge S. Marques, Segmentation and Classification of Human Activities, Proc. of HAREM2005 - Workshop on Human Activity Recognition and Modelling, held at BMVC2005 - British Machine Vision Conference, pp. 79-86, Oxford, UK, 2005
[9] Pedro Jorge, Arnaldo J. Abrantes, Jorge S. Marques, Estimation of the Bayesian Network Architecture for Object Tracking in Video Sequences, International Conference on Pattern Recognition, vol. 2, 732-735, 2004
[10] Jorge Silva, Jorge S. Marques, J. M. Lemos, Trajectory Tracking in Manifolds with Tangent Bundle Approximation, Proc. NOLCOS 2004 - 6th IFAC Symposium on Nonlinear Control Systems, Stuttgart, Germany, 2004
[11] Pedro Jorge, Arnaldo J. Abrantes, Jorge S. Marques, On-line Tracking Groups of Pedestrians with Bayesian Networks, Workshop PETS, 65-72, Prague, Czech Republic, 2004
[12] Pedro Jorge, Arnaldo J. Abrantes, Jorge S. Marques, On-line Object Tracking with Bayesian Networks, Proc. WIAMIS 2004 - International Workshop on Image Analysis for Multimedia Interactive Systems, Lisbon, Portugal, 2004 - PDF
[13] Jacinto Nascimento, New Performance Evaluation Metrics for Object Detection Algorithms, Workshop PETS, Prague, Czech Republic, 2004
[14] Arnaldo J. Abrantes, Jorge S. Marques, Joao M. Lemos, Long Term Tracking Using Bayesian Networks, IEEE International Conference on Image Processing, Rochester, 609-612, vol. III, 2002 - PDF
[15] Jorge S. Marques, Pedro Jorge, Arnaldo J. Abrantes, J. M. Lemos, Tracking Groups of Pedestrians in Video Sequences, Proc. WOMOT2003 - IEEE Workshop on Multi-Object Tracking, Madison, Wisconsin, USA, 2003 - PDF
[16] Jacinto Nascimento, Jorge S. Marques, João Sanches, Estimation of Cardiac Phases in Echographic Images Using Multiple Models, Proc. ICIP2003 - The IEEE International Conference on Image Processing, Barcelona, Spain, 2003 - PDF
[17] Jorge Silva, Jorge S. Marques, Joao M. Lemos, A geometric approach to motion tracking in manifolds, Proc. SYSID2003 - 13th IFAC Symposium on System Identification, Rotterdam, The Netherlands, 2003 - PDF
[18] Jacinto Nascimento, Arnaldo J. Abrantes, Jorge S. Marques, The Role of Middle level features for robust shape tracking, 12th Portuguese Conference on Pattern Recognition, Aveiro, 2002 - PDF
[19] Jorge Silva, Jorge S. Marques, J. M. Lemos, Selecting Landmarkpoints for Sparse Manifold Learning, Proc. of NIPS2005 - Neural Information Processing Systems Conference, Whistler, British Columbia, Canada, 2005