Zdenek Kalal

http://www.yovae.com/blog/zdenek-kalal%e8%b7%9f%e8%b8%aa%e7%ae%97%e6%b3%95.html

 

Zdenek Kalal是英国萨里大学的一个捷克学生。他演示的是他的神奇的精确定位系统，这个系统几乎可以跟踪镜头里的任何物体，只要你能看见它，并把它选中。它能 做很多神情的事情。在这个视频中，他演示了通过摄像机拍摄他的手指、把他的手指选做目标。系统于是就能精确的跟踪他的手指的动作。更令人惊奇的是，这个系 统能够通过分析物体的运动来完善跟踪算法。你能在很短的时间里教会它跟踪你的手指、面孔或在高速公路上狂颠的轿车。有了这套系统，我们几乎真的可以实 现”Minority Report“那样的人机界面。就像微软Xbox的Kinect那样，而这个效果更好。

Kalal有12个视频来演示 他的这套算法都能做什么。只要你有一个好的摄像头，把这个软件装到计算机上、平板电脑上或手机里，它就能精确的定位跟踪你的前额上的一个点、你的指尖、或 你的眼睛。你把摄像头放到门外，它就能自动识别是你认识的人来了，或警告你这是个陌生人。人们不用通过手就能简单的操控计算机。这项技术应用前景广泛。

你可以从萨里大学的网站找到这个程序的代码，它是免费的。Kalal被授予了“Technology Everywhere”奖学金作为嘉奖。来自: 外刊IT评论

—–

TLD has been developed by Zdenek Kalal during his PhD thesis supervised by Dr. Krystian Mikolajczyk and Prof. Jiri Matas. The main contributions of TLD have been presented at international computer-vision conferences where TLD tracker significantly outperformed state-of-the-art approaches. For his work on TLD, Zdenek Kalal has been awarded the the UK ICT Pioneers 2011 prize.

Key Features

• Single object tracking
• No offline training stage
• Real-time performance on QVGA images
• Implementation: Matlab + C, single thread, no GPU
• Dependence on OpenCV library (single function)
• Ported to Windows, Mac and Linux
• Illumination invariant features

More Information

Poster, Papers: ICCV’09 (w), CVPR’10, ICIP’10, ICPR’10

The publications listed on that page are:

• [5] Z. Kalal, K. Mikolajczyk, and J. Matas, “Face-TLD: Tracking-Learning-Detection Applied to Faces,” International Conference on Image Processing, 2010.
  [pdf][poster][ 
  ]
• [4] Z. Kalal, K. Mikolajczyk, and J. Matas, “Forward-Backward Error: Automatic Detection of Tracking Failures,” International Conference on Pattern Recognition, 2010, pp. 23-26.
  [pdf][ 
  ]
• [3] Z. Kalal, J. Matas, and K. Mikolajczyk, “P-N Learning: Bootstrapping Binary Classifiers by Structural Constraints,” Conference on Computer Vision and Pattern Recognition, 2010.
  [pdf][poster 1][poster 2][ 
  ]
• [2] Z. Kalal, J. Matas, and K. Mikolajczyk, “Online learning of robust object detectors during unstable tracking,” On-line Learning for Computer Vision Workshop, 2009.
  [pdf] [ 
  ]

Free Version

TLD can be downloaded for testing in a chosen application. We provide a precompiled demo (Windows) and a source code that is released under GPL version 3.0. In short, it means that any distributed project that includes or links any portion of TLD source code has to be released with the source code under the GPL version 3.0 license or later.