Change in the direction

Recently, I made a decision to remove the (Ubuntu) Linux from my notebook as a native operating system. I'm just tired about the troubles with the drivers etc. I did not throw the development of Cognitive Vision away, now my main platform will be the Windows, but I keep in practice to use open-source tools and libraries. Currently, the build system of myLibrary is changed from autotools to cmake and I can compile and run everything both Windows and Ubuntu Linux (in VM).

Felmondtam

A mai napon felmondtam a jelenlegi munkahelyemen, és már csak idő kérdése, hogy aláírjam a szerződésemet az új helyen. Nagyon jó lehetőséget kaptam. A felmondási időm 2 hónap...

Motion detection with moving camera

I expanded my motion detection algorithm in order to work with moving cameras. It was not so difficult, but I spend last weeks with enhancements to be better and better. Now it works quite fine, but I'm implementing a few new additions to the motion detection to be more robust. Current results will come in the new version of my paper.

Új kabát

A tegnapi napon hazafelé menet sikeresen sikerült vennem egy elég igényes kabátot, s mint a pénztárnál kiderült akciós, úgyhogy majdnem az eredeti ár feléért vittem haza. :)

Turkui húsvét

A hétvégét a családommal Turkuban töltöttük a kedvesem egyik ismerős családjánál. Eléggé jó volt, bár a kabátom tönkrement, amit szerettem, még Mo.-ról volt, úgyhogy a héten venni kell egy újat... A nyuszik többször is megajándékozták a gyerekeket csokival, illetve nagyon jókat ettünk a vendéglátó brazil családnál. :)

My paper

Unfortunately, my paper got quite bad opinions in the review process of the conference where I sent. I was a bit disappointed, because I had a lots of work in that paper and I did not expect so wrong result. After some depression, I got the point, that the review comments are very helpful: this paper is out of scope of the conference, I should rewrite some parts of the paper and put some missing things, which are important from point of view of the researchers/reviewers who reads it.
My English is poor, it was mentioned in more places, but I don't think that I will be an "English master", thus it can remain a weak point. Let's work on enhancements!

Image stacking

I tried image stacking in FindShapes, but it does not seem to be efficient in case of dynamic object recognition applicition, thus it is dropped...