I know Horn and Schunck and Lucas-Kanade (81) as optic flow methods. But Nagel is way to complex for me. Anyway I have an blackbox implementation of this method but would like to understand the improvements compared to the two basic methods for optic flow computation.

They calculate higher derivations and they seem to look at more than one frame ahead in time. But I am still not able to evaluate how the algorithm works because the paper is way to complicated for me.

Can somebody tell me what exactly the advantages and disadvantages of this method are?

Thank you in advance.

  • 2
    $\begingroup$ I think, I would be more helpful, if you can provide your understanding of the algorithm (like in a stepwise manner) and ask for parts that you couldn't figure out. $\endgroup$ – MimSaad Aug 21 '17 at 13:39
  • $\begingroup$ My problem is that I don't get the algorithm and don't understand the paper (I am no mathematician). I just want to know how the algorithm works. Not the details but a coarse overview of the improvements. Compared to basic methods - what is used and how. I do not expect to understand the paper and way how they created the algorithm. But I want to understand the characteristics of the algorithm. $\endgroup$ – baxbear Aug 21 '17 at 13:51
  • 1
    $\begingroup$ I meant what was your efforts to understand the algorithm. Add those to your question. Which part exactly is unclear. For example, how it introduces sampling constrains, why it works on 3 consecutive frames, etc. $\endgroup$ – MimSaad Aug 21 '17 at 15:29
  • $\begingroup$ github.com/fraudies/optic-flow-estimation $\endgroup$ – MimSaad Aug 21 '17 at 15:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.