[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-17 Thread T. Modes
> > Calling autopano-sift-c multiple times will result in a lot of > > duplicated work, probably what is needed is to go back to the old > > autopano-c-complete.sh script which will cache features as .key > > files between runs. > I implemented also the two step approach. Now for multi-row panoram

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread T. Modes
Hi Terry and Bruno, > >> I'm pretty sure this is not how it is supposed to work.  Thomas has > >> actually reimplemented all the logic from gigastart within Hugin, > >> i.e.  what you enter for Program and Arguments are settings for a > >> 'normal' control point detector such as autopano-sift-c or

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Tduell
Hullo Bruno/All, On Jan 13, 10:31 am, Bruno Postle wrote: [snip] > gigastart creates a vast number of temporary files, but these are > reused by the Makefile system, so images are only opened once for > classifying features (though looking at the code, I don't see how > gigastart would work fro

Re: [hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Bruno Postle
On Tue 12-Jan-2010 at 14:36 -0800, Terry Duell wrote: On Jan 13, 9:16 am, Bruno Postle wrote: I'm pretty sure this is not how it is supposed to work.  Thomas has actually reimplemented all the logic from gigastart within Hugin, i.e.  what you enter for Program and Arguments are settings for a

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Tduell
Hullo Bruno/All, On Jan 13, 9:16 am, Bruno Postle wrote: > On Tue 12-Jan-2010 at 13:58 -0800, Terry Duell wrote: > > > > > > >On Jan 11, 6:04 pm, "T. Modes" wrote: > > >> I implemented this algorithm inside hugin. To use it please create a > >> new cp detector setting and set type to multi-row.

Re: [hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Bruno Postle
On Tue 12-Jan-2010 at 13:58 -0800, Terry Duell wrote: On Jan 11, 6:04 pm, "T. Modes" wrote: I implemented this algorithm inside hugin. To use it please create a new cp detector setting and set type to multi-row. I added also a variant for stacked projects. I tried this (trunk svn 4880), usin

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Tduell
Hullo Thomas/All, On Jan 11, 6:04 pm, "T. Modes" wrote: > I implemented this algorithm inside hugin. To use it please create a > new cp detector setting and set type to multi-row. > I added also a variant for stacked projects. I tried this (trunk svn 4880), using the simple multi-row set of pho

Re: [hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-12 Thread Oskar Sander
Sounds interesting Thomas. ould it do all thos setps as a part of the "align" then you mean? Is this Trunk? /Cheers 2010/1/11 T. Modes > I implemented this algorithm inside hugin. To use it please create a > new cp detector setting and set type to multi-row. > I added also a variant for stac

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-10 Thread T. Modes
> So here is my recipe for efficiently matching a multi-row panorama: > > 1. First use ptochain to connect all consecutive photos in the > project, the result will be a project with one or more groups of > connected photos - With luck each of these groups will correspond > with a 'row' of the panor

[hugin-ptx] Re: large multi-row panoramas (was:Makefile generation)

2010-01-07 Thread Pablo d'Angelo
Hi Bruno, Bruno Postle wrote: > So here is my recipe for efficiently matching a multi-row panorama: > 1. First use ptochain to connect all consecutive photos in the project, > the result will be a project with one or more groups of connected photos > - With luck each of these groups will correspo