Many of the problems in Image processing are embaressingly parallel. But I want to know is it worth solving Image processing using parallel computing? Are there real practical problems in image processing that can benefit form parallel computing? If so what are those? I am not taking about mliticore CPUs, instead I am talking about parallelism offered by GPUs, and clusters of GPUs.


closed as too broad by Christian Clason, Jan, Brian Borchers, Max Hutchinson, Godric Seer Oct 28 '13 at 19:34

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ I guess it depends on the problem, like what kind of algorithm you are using, what specific problem you are solving, or maybe the language you are using. if the processing is explicit of course it can be paralleled easily. the question should be more specific. $\endgroup$ – Chenming Zhang Oct 23 '13 at 6:47
  • 1
    $\begingroup$ I think this question is too broad for a useful answer, but a short reply would be: Yes, in particular if you want to do real time processing of image sequences. Have a look at some of the things on gpu4vision.icg.tugraz.at. $\endgroup$ – Christian Clason Oct 24 '13 at 14:47
  • $\begingroup$ Ah! The days using 50+ cores on the University's cluster to calculate the optical flow on 100,000+ images :) Would have taken 3 months at home... $\endgroup$ – Damien Oct 24 '13 at 21:35

Browse other questions tagged or ask your own question.