Hi Steve, It sounds like you want to write a python extension - a compiled python module that has access to the DD::Image classes... I'm guessing this is possible, in fact I think the PythonGeo.cpp example demonstrates this, but i have only glanced at the code.
Colin On 13 July 2012 09:13, Stephen Newbold <[email protected]> wrote: > Thanks Howard, yeah I have written a plugin that can do this all live and > fast but this needs to be python as it needs to be embedded inside another > node. nuke.sample() is slow but seems to be the only way to go so I'll > probably have to not run in live and treat it as something that needs to > analyse a sequence like other slow tools. > > Cheers, > Steve > > Howard Jones wrote: > >> Ben Pierre wrote this - >> http://www.nukepedia.com/**plugins/colour/imagetool/<http://www.nukepedia.com/plugins/colour/imagetool/> >> >> works live as well but wont return pixel position >> Howard >> >> ------------------------------**------------------------------** >> ------------ >> *From:* Stephen Newbold <[email protected]> >> *To:* Nuke Python discussion >> <nuke-python@support.**thefoundry.co.uk<[email protected]> >> > >> *Sent:* Thursday, 12 July 2012, 15:00 >> *Subject:* Re: [Nuke-python] Max velocity from a scan line >> >> >> Hmm, not really as I say it needs to be in a self-contained python >> script embedded in the scanline renderer itself. You can get a >> decent 'live' min/max using the dilate/erode trick or just write a >> custom plugin to do this nice and fast but I can't go there for this. >> >> Steve >> >> Hugo Léveillé wrote: >> > Lazy answer but would using to curve tool to extract max data be an >> > option instead of looping pixels? >> > >> > On Thu, Jul 12, 2012, at 13:46, Stephen Newbold wrote: >> > >> Hi, I have a tool that needs to obtain the maximum velocity >> value from a scanline renderer. Currently I am finding the max >> pixel value of the motion channel generated by the scanline >> renderer by iterating through each pixel of the motion channel and >> returning the max(). A basic but inaccurate optimisation is to >> skip x pixels to speed this up but I was hoping there might be a >> more cunning way of doing this? >> >> >> >> I'm presuming the Scanline Renderer itself doesn't store >> min/max velocities in a hidden knob or something else accessible? >> >> >> >> I was hoping to keep this process live and contained in an >> embedded python script, so using NDK or a node tree (dilate to >> value of width of image worked well enough if I went this method) >> so any suggestions are welcome. >> >> >> >> Steve >> >> >> >> -- Stephen Newbold >> >> Compositing Lead - Film >> >> MPC >> >> 127 Wardour Street >> >> Soho, London, W1F 0NL >> >> Main - + 44 (0) 20 7434 3100 >> >> www.moving-picture.com >> >> >> >> ______________________________**_________________ >> >> Nuke-python mailing list >> >> >> Nuke-python@support.**thefoundry.co.uk<[email protected]> >> >> <mailto:Nuke-python@support.**thefoundry.co.uk<[email protected]> >> >, >> >> http://forums.thefoundry.co.**uk/ <http://forums.thefoundry.co.uk/> >> >> >> http://support.thefoundry.co.**uk/cgi-bin/mailman/listinfo/** >> nuke-python<http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python> >> >> > >> > >> > >> >> -- Stephen Newbold >> Compositing Lead - Film >> MPC >> 127 Wardour Street >> Soho, London, W1F 0NL >> Main - + 44 (0) 20 7434 3100 >> www.moving-picture.com >> >> ______________________________**_________________ >> Nuke-python mailing list >> >> Nuke-python@support.**thefoundry.co.uk<[email protected]> >> >> <mailto:Nuke-python@support.**thefoundry.co.uk<[email protected]> >> >, >> http://forums.thefoundry.co.**uk/ <http://forums.thefoundry.co.uk/> >> http://support.thefoundry.co.**uk/cgi-bin/mailman/listinfo/** >> nuke-python<http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python> >> >> >> ------------------------------**------------------------------** >> ------------ >> >> >> ______________________________**_________________ >> Nuke-python mailing list >> Nuke-python@support.**thefoundry.co.uk<[email protected]>, >> http://forums.thefoundry.co.**uk/ <http://forums.thefoundry.co.uk/> >> http://support.thefoundry.co.**uk/cgi-bin/mailman/listinfo/**nuke-python<http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python> >> >> > > > -- > Stephen Newbold > Compositing Lead - Film > MPC > 127 Wardour Street > Soho, London, W1F 0NL > Main - + 44 (0) 20 7434 3100 > www.moving-picture.com > > ______________________________**_________________ > Nuke-python mailing list > Nuke-python@support.**thefoundry.co.uk<[email protected]>, > http://forums.thefoundry.co.**uk/ <http://forums.thefoundry.co.uk/> > http://support.thefoundry.co.**uk/cgi-bin/mailman/listinfo/**nuke-python<http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python> > -- colin alway
_______________________________________________ Nuke-python mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-python
