I'm going to be working on this project again in the near future. The current plans are to adapt the algorithm for video data. This will probably involve considering motion and various other things, and maybe it will even get optimized or something fancier than that.
The goal will be to implement this on a GPU (in CUDA and/or OpenCL) so that it can handle high definition video streams in as close to real time as possible.
Hi,
ReplyDeleteGreat job! Congratulations! :-)
Do you have any ETA already for a CUDA implementation? We're doing work on Bayesian multimodal perception and we've also done stuff on CUDA to get some speedup (see http://www.springerlink.com/content/e220025575672854/), and we're very interested in reusing your code (with respective citing, of course!)...
Cheers from Portugal
We should have the implementation finished within a few weeks and I will make a post about it here. It's not quite the same algorithm as documented on this website, but all code will be posted when it is finished.
ReplyDeleteHi,
ReplyDeleteVery interesting work. Is there any news on when the CUDA version will be available? Would like to compare its performance with a couple of other visual systems, with a possibility to use it for our robotic platform.
Cheers