Re: [hugin-ptx] Google Summer of Code 2013

2013-02-27 Thread Robert Campbell
On Feb 21, 2013, at 7:09 PM, Jim Watters wrote: Hugin and Panotools community, What new ideas do you have that a student could implement? I'm the type to make panos while holding my p-n-s Canon S95. I've tried doing some bracketed shots and create HDR from that, but I've had limited

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-27 Thread JohnPW
I'm not following this idea. Can you elaborate a bit? How would a control dowel work and what would it do? On Wednesday, February 27, 2013 9:48:52 PM UTC-6, Bob Campbell wrote: On Feb 21, 2013, at 7:09 PM, Jim Watters wrote: Hugin and Panotools community, What new ideas do you have

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-27 Thread Tduell
Hello Bob, It seems to me that you really only need two dowels, probably best not too close to one another. The first dowel would align all of the first control points, and the second align all the second control points thus ensuring all images are rotated around the first dowel correctly.

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-27 Thread Gnome Nomad
Like a control dowel would be a line (defined as straight) that connects single points between multiple separate frames? An any angle generalization of a horizontal line spanning multiple frames? My understanding of horizontal lines is probably flawed ... On 02/27/2013 05:48 PM, Robert

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-27 Thread Robert Campbell
Yes, this and what Gnome summarized are close to what I was thinking. So, if you have two adjacent stacks of three exposures, presumably they would all overlap by at least 10% and all six images would share a particular control point - thus creating a single line down through both stacks.

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-24 Thread Bruno Postle
On Sat 23-Feb-2013 at 22:35 -0800, Christoph Spiel wrote: Am Freitag, 22. Februar 2013 13:32:38 UTC+1 schrieb Bruno Postle: Merging nona with enblend/multiblend. Currently we remap each photo with nona, then enblend/enfuse splits these remapped images into multilevel pyramids, joins them into a

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-23 Thread Monkey
Merging nona with enblend/multiblend... Another way of doing this would be to split the input photos into pyramids, then remap each directly into the final pyramid.. I have a feeling - just a feeling at the moment, I haven't done any tests - that the operations can't really be swapped. I

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-23 Thread Bruno Postle
On Sat 23-Feb-2013 at 09:25 -0800, David Horman wrote: and I suspect doing it like this would avoid the nasty 'whorl' artefacts we get at the zenith/nadir (this assumption needs testing). Do you have an example of the whorls? You see the 'whorls' when you stitch to equirectangular,

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-23 Thread cspiel
Am Freitag, 22. Februar 2013 13:32:38 UTC+1 schrieb Bruno Postle: Merging nona with enblend/multiblend. Currently we remap each photo with nona, then enblend/enfuse splits these remapped images into multilevel pyramids, joins them into a single pyramid which is then reassembled into a final

Re: [hugin-ptx] Google Summer of Code 2013

2013-02-22 Thread Bruno Postle
I'm definitely not going to have time for mentoring the summer of code this year. This is a short list of features that are in Panotools::Script that I think ought to be moved into Hugin, maybe one of them could be the basis of a project: Adding GPano XMP tags to all equirectangular output.

[hugin-ptx] Google Summer of Code 2013

2013-02-21 Thread Jim Watters
Hugin and Panotools community, The Google Summer of Code 2013 is on. Where Google funds a student to work on Free Software like Hugin for the summer. http://google-opensource.blogspot.ca/2013/02/flip-bits-not-burgers-google-summer-of.html