April 4, 2011 at 18:10Tracking objects on video
A Czech student from the University of Surrey, UK, Zdenek Kalal, as part of the practical part of the Ph.D. thesis, developed the Tracking-Learning-Detection (aka Predator) algorithm for tracking objects in a self-learning video stream (recognition accuracy improves with each frame).
Demo of the project
Source codes on github: 1 , 2 , 3 , 4 , 5
The program works normally in a single stream of Intel Core 2 Duo 2.4 GHz, 2 GB RAM.
On the Zdenek page on Youtube, you can find other videos.
The author says that he was going to publish the source codes of the project, but when he received hundreds of letters from interested parties asking him to do this, he changed his mind. It turned out that this project is much more interesting than it was originally intended, so there was a chance of earning something.
Such algorithms can be used in tracking systems (for example, for unmanned vehicles), computer and game interfaces (virtual mouse), in photo and video cameras, etc.
In fact, this technology is unlikely to be as unique as the author believes. As far as I know, Russian developers from Intel R&D in Nizhny Novgorod do something similar (at a conference Intel showed a demo of their system).
See also:
Details on Breakthrough AI in Kinect
UPD 04/05/2011 12:23. The article had an effect and Russian developers also moved. This alogirtm from German Bukharov does not look so impressive, but it can even work on smartphones.
UPD2 04/05/2011 18:14. Computer vision system from the company Rhonda Software (Vladivostok). via
Demo of the project
Source codes on github: 1 , 2 , 3 , 4 , 5
The program works normally in a single stream of Intel Core 2 Duo 2.4 GHz, 2 GB RAM.
On the Zdenek page on Youtube, you can find other videos.
The author says that he was going to publish the source codes of the project, but when he received hundreds of letters from interested parties asking him to do this, he changed his mind. It turned out that this project is much more interesting than it was originally intended, so there was a chance of earning something.
Such algorithms can be used in tracking systems (for example, for unmanned vehicles), computer and game interfaces (virtual mouse), in photo and video cameras, etc.
In fact, this technology is unlikely to be as unique as the author believes. As far as I know, Russian developers from Intel R&D in Nizhny Novgorod do something similar (at a conference Intel showed a demo of their system).
See also:
Details on Breakthrough AI in Kinect
UPD 04/05/2011 12:23. The article had an effect and Russian developers also moved. This alogirtm from German Bukharov does not look so impressive, but it can even work on smartphones.
UPD2 04/05/2011 18:14. Computer vision system from the company Rhonda Software (Vladivostok). via