%0 Conference Paper %1 SmartZhang:04:ivcnz %A Smart, Will %A Zhang, Mengjie %B Proceeding of Image and Vision Computing International Conference %C Akaroa, New Zealand %D 2004 %E Pairman, David %E North, Heather %E McNeill, Stephen %I Lincoln, Landcare Research %K Artificial Computer Intelligence, Object Tracking, Vision algorithms, genetic programming, %P 113--118 %T Tracking Object Positions in Real-time Video using Genetic Programming %U http://www.mcs.vuw.ac.nz/~mengjie/papers/1051-will-meng-ivcnz04.pdf %X Genetic Programming (GP) to evolve programs for tracking objects quickly in streaming video. A small number of images, with located objects, are used as training data and GP automatically performs feature-selection on these images at the pixel level. The use of feature functions is introduced, taking a single offset argument, in contrast to the standard feature terminal approach. The features include both 'directionless' intensity features and 'directional' edge detection features. The fitness function rewards evolved programs that can move training points, located on a grid around an object, closer to the object. As such, a good program will also be able to update an object position from frame to frame for tracking. Two video sequences are examined, with evolved programs tracking the left-eye and forehead of a person successfully. The method is very fast, tracking a frame in six or seven milliseconds on a 2.6GHz PC.

@inproceedings{SmartZhang:04:ivcnz, abstract = {Genetic Programming (GP) to evolve programs for tracking objects quickly in streaming video. A small number of images, with located objects, are used as training data and GP automatically performs feature-selection on these images at the pixel level. The use of feature functions is introduced, taking a single offset argument, in contrast to the standard feature terminal approach. The features include both 'directionless' intensity features and 'directional' edge detection features. The fitness function rewards evolved programs that can move training points, located on a grid around an object, closer to the object. As such, a good program will also be able to update an object position from frame to frame for tracking. Two video sequences are examined, with evolved programs tracking the left-eye and forehead of a person successfully. The method is very fast, tracking a frame in six or seven milliseconds on a 2.6GHz PC.}, added-at = {2008-06-19T17:35:00.000+0200}, address = {Akaroa, New Zealand}, author = {Smart, Will and Zhang, Mengjie}, biburl = {https://www.bibsonomy.org/bibtex/2e8ecb38f32711d54ee274625c69c405a/brazovayeye}, booktitle = {Proceeding of Image and Vision Computing International Conference}, editor = {Pairman, David and North, Heather and McNeill, Stephen}, interhash = {fc5fd0f6af35edb648ab0a076a471b63}, intrahash = {e8ecb38f32711d54ee274625c69c405a}, keywords = {Artificial Computer Intelligence, Object Tracking, Vision algorithms, genetic programming,}, month = {November}, notes = {Fri, 02 Jun 2006 17:03:20 +0800}, pages = {113--118}, publisher = {Lincoln, Landcare Research}, timestamp = {2008-06-19T17:51:48.000+0200}, title = {Tracking Object Positions in Real-time Video using Genetic Programming}, url = {http://www.mcs.vuw.ac.nz/~mengjie/papers/1051-will-meng-ivcnz04.pdf}, year = 2004 }