FBK home > INFORMATION TECHNOLOGY > Technologies of Vision > dynamic scene understanding

Technologies of Vision: dynamic scene understanding

SmarTrack - an adaptive system for real-time tracking of multiple people in videos

Reference project - FBK Joint Research Project PUMALAB - Attentive Bayesian Perception using Multi-modality, Adaptation and Learning


a frame

* description | potential application areas | achievements

* people tracking - mpeg video (11 669 504) illustrating how SmarTrack works

* SmarTrack@IST2006 - SmarTrack awarded one of three winners of the Best IST 2006 Exhibit Award.

* CHIL EU Project (IST-2002-506909) - CHIL Technologies: SmarTrack [pdf]

* FBK SpotLight n. 15/2009 FBK SmarTrack video [wmv]


 

 

 

US Patent 7965867  -  EU Patent Application EP20060116896 pending (06116896.9)
Method and Apparatus for Tracking a Number of Objects or Object Parts in Image Sequences

EU Patent 2302589  -  US Patent 8436913
Method for Efficient Target Detection from Images robust to Occlusion

 

References

[1] Oswald Lanz, "Approximate Bayesian Multibody Tracking", IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(9):1436-1449, September, 2006 [pdf]

[2] Oswald Lanz, Roberto Brunelli, "Dynamic Head Location and Pose from Video", IEEE Conference on Multisensor Fusion and Integration, September 3-6, 2006, Heidelberg, Germany

[3] Oswald Lanz, "An Information Theoretic Rule for Sample Size Adaptation in Particle Filtering", International Conference on Image Analysis and Processing - ICIAP 2007, September 10-14, 2007, Modena, Italy

[4] Oswald Lanz, "A HJS Filter to Track Visually Interacting Targets", IEEE International Conference on Acoustics, Speech, and Signal Processing - ICASSP 2009, Taipei, Taiwan, April 19-24, 2009

[5] Oswald Lanz, Stefano Messelodi, "A sampling algorithm for occlusion robust multi target detection", IEEE International Conference on Advanced Video and Signal based Surveillance - AVSS 2009, Genova, Italy, September 2-4, 2009

SmarTrack - info

Please contact
Oswald Lanz  |  e-mail lanz (at) fbk . eu