So what exactly is the focus of your research?
aeh...compression stuff. Actually my thesis is gonna be a big project about compression for still images and video. Right now im specializing in compressing CMYK images that are used by printers for printing images and its sponsored by some big printer companies, like one that I cant remember the name of (laughs), but its one of the biggest in the US. I had never heard of it before, but its one of the biggest in the US - the other is Epson.
Thats the first part of my thesis, expanding into everything that needs to be compressed: Video, we already have a very good alogrithm for RGB images from previous guys that worked with Dr. Yfantis, so we kind of expand to everything, every kind of still images, with CMYK now, because these are the two main formats in internal image representation. And then we are gonna do video! 

	So you apply the algorithm to all kinds of compression?
You use the same mathematical tools, but the algorithms are very different when you move to moving images. 

	What are you working on at the moment?
I experiment on two images, the one is in Encapsulated Postscript Format, the other is in CMYK which is more than 100 megs. It took me more than 20 minutes to load it in Photoshop and I can view it in my viewer in less than 3 minutes. 

	How did you get into writing a viewer for CMYK images?
I will make a viewer for the things I learned because I did not do anything in image processing before. I did theoretical stuff in Greece - I will just give it away. Its gonna be a smart viewer because even big products, viewers like Photoshop, I donīt think they can read very large images in clever way. I will do one of these things mainly for my research, because I need to see those large images quickly, and most probably I will give it away. 

	What is the difference between a regular viewer and your program?
How it does it? It reads the header, finds the height and the width and allocates the whole memory, then just starts reading serial - I do it kind of parallel because one thread is reading or two threads are reading or three, i dont know because I decide on the fly and the other threads are converting to RGB at the same time. GSview took about 25 minutes to read a CMYK TIFF image of 130 megs. It took me three and a half minutes. 

	So it will be possible to download the viewer?
Its ready, but it still needs some bells and whistles, some interface stuff. 

	What was your motivation to come to Las Vegas and do your master degree here?
There is a very good research team with Dr. Yfantis, many cooperations with the industry, the computer industry. In Greece I did lots of stuff that no one will use ever except a few of them, here you see the result of your work, people use it. It is a nice a feeling.
In Europe the progress is much slower. People are used to a different working style. And you know, when you migrate to another country you have no friends, its easier to focus, you know, to spend all your time on that. You dont have friends coming over to your house, no parties, no family. 

	So there is also a downside?
There is always a trade-off, you know. That I work alot and dont have much spare time - thats the only thing, the only bad thing. 

	It was a pleasure talking to you and we are looking forward to the next time.



<__b.a.c.k.______