[jmzachary]

Interesting. CBIR was my dissertation topic and I haven't thought about it in some time. Brings back fond memories (and some kooky thoughts about what to with it now.) Thanks, nickb.

It's a hard problem (computationally and figuratively) since perfect image search is basically the same problem as computer vision. Most people think of image search like they do textual search: exactness is a requirement. I relaxed that requirement and look at different methods to cull the search space down (throw out images that probably weren't relevant).

[nickb]

jmzachary, you're welcome! Would you recommend me some resources on CBIR?

Thanks in advance!