Monthly Archives: September 2010

Modifications : The original package of LIBSVM 3.0 was little bit modified to include intersection and Chi-square kernels. Intersection kernel is particularly useful for histograms and has been shown to be efficient in computer vision tasks. Chi-square kernel was shown … Continue reading →

Hey, i just built and successfully compiled my first ever Qt application on a Mac! Yay! Here is the sample code : #include <QtGui> int main(int argc, char *argv[]) { QApplication app(argc, argv); QLabel label(“Hello, world!”); label.show(); return app.exec(); } … Continue reading →

Who has tried to build some Qt project using Cmake under Mac OS X Snow Leopard (10.6)? I believe that you may get this error : “You are building a 64-bit application, but using a 32-bit version of Qt.” What … Continue reading →

So, how in the world you are supposed to convert full directory of JPEGs say to PPMs? Open each and “Save as..” separately? < — i worry about your spirit about this operation Make a script and use kind a … Continue reading →

Has anybody thought what the word “Universe” means? Uni = One Verse = Sentence So, all we have is coming from one single sentence… from the God!

I have an idea how, for example, USB removable media should worked. Assume that you quite often use your USB flash to transfer files and you find that you spend annoying moments going to eject/unmount/safe remove/etc your drive. What i … Continue reading →

Check out videos from Kent Hovind … and be yourself when watching them. Debates — wish to see a good battle of words? Then Kent’s talk at Oxford is for you! You won’t regret your time spend! Some free seminars … Continue reading →

Nice package for decision function learning in machine learning domain… well, it can possibly help you out to classify the proposed house you wish to by (given price, m2, location etc) as good or not. It can be downloaded from … Continue reading →

This paper covers the problematic of video indexing domain. Given first few minutes annotated by human user, the goal is to propagate the labels to the parts of video without them. The paper proposes semi-supervised learning approach which leverages both … Continue reading →