I need to decide what topic I would like to research for my master thesis. I really like to thinker with OpenGL development, so I know I will do something related, but it’s been a long time since I wrote last semi advanced line of code in OpenGL, so now I got to catch up.
Last thing I did was a demo that was able to do reflection, refraction stuff, plus it could switch between bump/paralax/relief mapping techniques and could do all of that in FP16 NPOT off screen buffers to enable me to get some histogram data and do some minimal post processing, most of it in GLSL.
Back in the day that was hot, but it’s been like two years since then, and now I have no idea what has changed.
Can someone suggest me what topics are interesting and solid enough in CG world now, so that I could decide about which topic I could write a thesis?
One that springs to mind is the ray tracing stuff that everyone is looking at now and the research involved in a move away from traditional raster GL.
Other than that I guess the new direction GL3.0 is going, and the ongoing evolution away from fixed to programmable pipelines, which can look at the advantages of instancing and Geometry Shaders.
Or what about OpenCL, although that’s not in the wild yet…
Depending on the university, they might not want you to do something too tightly coupled to OpenGL. The thesis would be on computer graphics and your implementation would be done using OpenGL.
Regardless, I agree with scratt. Shader based architectures are hot, so are geometry shaders. For example, in a GPU course I took a while back, one team implemented b-splines on the GPU using geometry shaders.
I’d also recommend looking through the recent SIGGRAPH and IEEE papers, especially the surveys. I must have scanned 100 papers before I even had an abstract. BTW my thesis is on out-of-core rendering. If that is something you are interested in, I can post a link once I get it online. Should be soon.
GPU Gems II is free online. It is a few years old but probably still worthwhile.
Also, Ke-Sen Huang’s Home Page has recent papers from just about every major graphics conference.
path finding (eg A*) on the GPU
Im sure lots of ppl will be very interested