Home arrow static arrow Java Programming [Archive] - Thesis Suggestions
Warning: Creating default object from empty value in /www/htdocs/w008deb8/wiki/components/com_staticxt/staticxt.php on line 51
Java Programming [Archive] - Thesis Suggestions
This topic has 5 replies on 1 page.

Posts:94
Registered: 2/18/04
Thesis Suggestions  
Jun 17, 2004 11:22 PM



 
Hello all. Good day.

The time has come for me to pick a thesis topic that I want to pursue, yet I am at a loss. At other universities, their theses consist of making decision support systems or information systems for schools or for companies. However, in our university, there's an added twist. The thesis must be "something new" or there must be an element that is "new" and we are expected to provide a working prototype to prove our concept. In other words, it must be something that hasn't been done before.

Ex: Grid Computing Using Cellphones

My professor has suggested that I pick a topic that interests me well, so that I will have the motivation to work on the project.

I am interested in the following subject areas: steganography, simulation, data compression and graphics. However, steganography seems saturated. Hiding information in image and audio files has already been done. The same goes for data compression. There are already a bunch of algorithms out in the open that can compress data by as much as 42% and above.

So what's left are simulation and graphics. I haven't thought much about simulation though, as graphics is a more interesting path. I had a look at Scalable Vector Graphics (SVG) and I was quite amazed at what it could do.

Could anyone give me some suggestions on the four subject areas that I mentioned? Maybe there's still something to be made in data compression or steganography. I want to keep my chances of finding a good thesis high, so suggestions from different areas of interest is good. I was thinking of a simple 3d modeler using SVG but I don't know if 3d SVG modelers have already been done.

Oh, and obviously, I want to do my thesis in Java. I've been wanting to develop on Java ever since we started (and finished) our software engineering class, which we spent with C#.NET. My experience with VS was "good, bored, then sick".

Cheers!
 

Posts:1,135
Registered: 1/16/04
Re: Thesis Suggestions  
Jun 17, 2004 11:50 PM (reply 1 of 5)



 
Could anyone give me some suggestions on the four subject areas that I mentioned?
steganography, simulation, data compression and graphics...

The thought that came to my mind was to combine the four... Create a simulation program that uses 3D graphics to represent the simulated item like say human locomotion where steganography is incorporated into the construction of the 3D model and data and transmits said models to a copy of your simulation program which has an encoder and decoder over a network using compression...

3D graphics model - contains stenographic data
simulation data - contains decryption\encryption key for stenographic data in model
data tranmission - transmits data over network with compression

Then again, maybe I need some sleep...

- MaxxDmg...

- ' He who never sleeps... '
 

Posts:94
Registered: 2/18/04
Re: Thesis Suggestions  
Jun 19, 2004 9:33 AM (reply 2 of 5)



 
Hey! Thanks for replying. Unfortunately, I think I wouldn't be able to complete such a thesis in a span of 10 (or 9) months. I could probably come up with a project that allowed me to reuse some of the existing packages for dealing with steganography and compression, but simulation would probably take up most of my time. Couple that with SVG, which I'm just starting to learn, and...

Since SVG is a graphics standard made for the web, I had this crazy idea of moving it to a client application without using a browser. Say for example, instead of icons on the desktop, they'd be replaced with SVG counterparts. Now whenever someone mouses over an SVG icon, the icon would do some cool animation stuff.

What do you think?
 

Posts:21,719
Registered: 98-02-20
Re: Thesis Suggestions  
Jun 19, 2004 9:51 AM (reply 3 of 5)



 
Something that's never been done before? That's a doctoral dissertation requirement, not an undergraduate thesis.

Pixar uses parallel processing algorithms to render movies like "A Bug's Life". What about looking into how that's done? Figuring out how to do it on a Linux cluster would not be trivial. I think it'd be interesting. Especially if you had one available to try it out. (Build one if you don't.)

What about a "physics in graphics" study to show how to realistically render something like fluid or hair motion efficiently? The hair on the donkey in "Shrek 2" was pretty amazing. Ever wonder how they do it? Maybe a survey of how fluid motion is done realistically without resorting to a full CFD simulation.

%

 

Posts:4,906
Registered: 23/07/02
Re: Thesis Suggestions  
Jun 21, 2004 10:49 AM (reply 4 of 5)



 
What level is this? BSc? MSc?
 

Posts:94
Registered: 2/18/04
Re: Thesis Suggestions  
Jun 22, 2004 10:19 AM (reply 5 of 5)



 
Sorry for not replying as quickly as I can, I had other things I needed to do...

What level is this? BSc? MSc?

It's for an undergraduate thesis, and if the BS you are referring to stands for Bachelor of Science, then that would be it.

I've asked around in our division and the thesis need not to be entirely new. It could be a better way of doing things (e.g. a better algorithm to triangulate a polygon).

duffymo's suggestions are interesting. They're obviously hard though. I will look into them when I have the time. Our teacher said that we can still change or suggest new topics. He ends that with a "topic close to your heart."
 
This topic has 5 replies on 1 page.