Anybody wants to write a Siggraph '07 paper with me?

NicoRi

Newcomer
Recently I graduated from university and wrote a paper about my final thesis (called diploma thesis over here in Germany). The paper has been accepted at Graphite 06 conference - a kind of (smaller) eastern version of the Siggraph. Now I wonder if it is worse trying to devolop my technique from the thesis further and write a paper about the result and submit it for the Siggraph 07. Anybody interested in joining? We would try to improve my technique significantly, so that there is enough content for a paper. IMO there is a good chance that this is possible.

In case you are interested, you find my Graphite paper, the thesis and a demo here. I'd be happy to find some interested and enthusiastic people to help me. We would have about four month time to improve the technique and write the paper. I think this is really possible. The deadline is about the 20th of January.

However, there is some uncertainty about how much time I will have during that 4 month because right now I'm unemployed, seeking a job. So in case I find a job I might not have much time for additional work left, but anyway I think it is worth a try! ;)

Nico

In case you are seriously interested, you can write me a private message - however I don't know yet how to send private messages myself... ;-)
 
Last edited by a moderator:
You could start by giving a short description of what the subject is about, why it is interresting, what it can be used for, etc ... That would increase chances that someone actually reads your paper.

But maybe that's just me.

Ozo.
 
Give him a break ... university students are used to papers downloading so fast that reading the abstract of the paper wouldn't really take more time than reading it on this thread ;)
 
You could start by giving a short description of what the subject is about, why it is interresting, what it can be used for, etc ... That would increase chances that someone actually reads your paper.

But maybe that's just me.

Ozo.

;)

Well, the subject is a technique to add detail to a (more or less coarse) polygonal mesh. You probably now relief mapping, hm? My approach is similar but uses a volume texture instead of a relief map or - as in the latest relief mapping incarnation - a layered relief map. This means my technique can handle true 3d geometry with arbitrary depth complexity. Ok, in practice the depth complexity is limited by the volume texture resolution. Anyway, another advantage over relief mapping is that my technique works in true 3d, i.e. there is a shell generated that sourrounds the polygonal mesh. The geometry volume texture is mapped into this shell by doing a simple ray tracing of view rays inside the shell space in the pixel shader. To skip empty space I use 3D distance map, so less search steps are requiered than in relief mapping and similar techniques to achieve good results. The results rendered so far with my technique look quite promising, however, the technique produces some nasty artifacts in some cases and also lacks a good filtering scheme (see paper and demo ;)). Also there is stuff I like to add like shadows i.e.. Anyway, got the idea?
 
It was an interesting demo. Although clearly there is work to be done :) (hence the post).
I found it odd it seemed to take a large performance hit from increasing the mesh complexity though.

As for Graphite, I don't know what your experience was like, but I was a bit dissapointed by graphite singapore. Some people were presenting to the stream coordinater only, while some (like me) started presenting to 1-2 people, and had 10+ by the end because the timing was so bad. Not that it mattered, I did an attrocious presentation :runaway:

Singapore is awesome very much :yes:
 
It was an interesting demo. Although clearly there is work to be done :) (hence the post).

Thanks a lot for the enthusiastic compliment, lol! ;-)

I found it odd it seemed to take a large performance hit from increasing the mesh complexity though.

This is because you have to adjust the number of ray tracing steps. The ray tracing accuracy automatically increases as you use more polygons.

As for Graphite, I don't know what your experience was like, but I was a bit dissapointed by graphite singapore. Some people were presenting to the stream coordinater only, while some (like me) started presenting to 1-2 people, and had 10+ by the end because the timing was so bad. Not that it mattered, I did an attrocious presentation :runaway:

Well, they seem to be quite disorganized. ;-) I also do not expect that many people to come...
but I don't know. What was the topic of your presentation? Did you ever wrote a Siggraph paper?

Singapore is awesome very much :yes:

Yep, I have been there too (not at Graphite). ;-)
 
I think your work is pretty cool. I have stumbled upon your work already, but I cannot remember where. I recall seeing those chain-linked meshes before.
 
;-) The mesh with the weaved wood pattern is from Fabio Policarpo, from Multi Layer Relief Mapping.
 
Yup. Included a comparison of the technique and mine in my paper...
Speaking of which... aren't you leaving it a bit late to start a Siggraph paper? They have to be of a very high standard (IIRC > 50% are rejected) and IIRC they have to be submitted in January.
 
Speaking of which... aren't you leaving it a bit late to start a Siggraph paper? They have to be of a very high standard (IIRC > 50% are rejected) and IIRC they have to be submitted in January.

Yes, you are right. That is why I was looking for people interested in writing a paper with me together. Time is running out and nobody has contacte me yet... Soon it will be to late for that project. Hm, maybe next year. ;-)
 
Back
Top