On 16 Jan., 23:25, Bruno Postle <[email protected]> wrote:
> Thanks for this, yes the bottleneck to distributed stitching is that > enblend has to assemble the final image and this requires a lot of > memory. > > One experiment we tried a long time ago was to split the enblend job > up, i.e. instead of blending images 0,1,2,3 in one go, you can blend > 0,1 and 2,3 as separate processes and then blend these together as a > third process. Just a far shot: if I understand the blending process correctly, the images are blended at different resolutions, so that low frequencies are treated differently to high frequencies. This would imply that the parts of the blending process which work at high resolution (high frequencies) don't need so much lateral context and could be parcelled off to be blended somewhere else, whereas the blending at lower frequencies could be done at reduced resolution locally, needing much less processing power. Splitting the images up with something like a DWT and then treating the different levels individually might be an option. Kay -- You received this message because you are subscribed to the Google Groups "Hugin and other free panoramic software" group. A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/hugin-ptx
