Summer of Code Students
Yuval Levy
already set up?
Yuv
Galloth
Well I was interested in proposal Straight-line detection for
automated lens calibration of your ideas. During my master studies I
took a few elective subjects about computer vision and computer
graphics and I wanted to test these knowledge with real word probem.
But I will not start programming until I will be selected, so I give
up on hugin (which requires sending a patches) although I understand
why you requiere it.
But when you asked, I am interested. Is that topic still free? Should
I write proposal even if I will work on hugin only after I got
selected?
Jan
2009/3/28, Yuval Levy <goo...@levy.ch>:
> anybody still around/interested other than León and Lukáš who are
> already set up?
>
> Yuv
>
> >
>
Jan Kastil
gal...@jabbim.cz
Yulia
Hi Yuv et al.,
I didn't want to post anything here before I finally set up everything. I spent quite a bit of time trying to compile the code on mac os. Instructions in the most cases are straightforward and easy to follow (there were a couple commands that are outdated. I can send a separate email later when I get to my machine). Right now the issue that I'm having is that when I try to stitch images together, it throws an error. I have a project for school due tomorrow morning, so I will get back to hugin after that. Would it be ok to ask for help with set up in this mailing list?
As for me, I'm a 4th year student at UofT studying AI. This term I got to take a course in Visual Computing and a grad course in Computer Graphics. Since I am really enjoying these subjects, I see hugin as an excellent opportunity to develop my skills in this area. For the past 2 years I was programming a lot in C++, I also have a decent knowledge of OpenGL and good mathematical background.
Currently I am very interested in improving the lens calibration for hugin. I recently studied couple of algorithms for precise lens calibration in one of my classes and would really like to work on implementing them.
I have two questions now: 1) does it matter which platform I am developing on? (at school I program on Mac and Linux only)
2) what level of detail do you expect to see in the proposal when it comes to implementation? (Obviously I will study the code as much as I can in the time left, but I cannot guarantee that I would be able to say precisely which data structures, functions etc. I plan to implement).
Sorry for longish email and thank you for your time.
Yulia
Yuval Levy
Yulia wrote:
> spent quite a bit of time trying to compile the code on mac os.
congratulations. that's not the easiest platform to deal with.
> Instructions
> in the most cases are straightforward and easy to follow (there were a
> couple commands that are outdated.
you're encouraged to sign up for a wiki account and update them.
> Right now the issue that I'm having is that when I try
> to stitch images together, it throws an error.
I'm not sure about the current status of SVN, and even less in relation
to building on OSX. Maybe Harry can recommend an SVN revision that is
known to work?
Other than that, errors are the daily bread of development. You're
invited to check if there is a similar report in the bug tracker and add
details there, or start a new ticket there.
<http://sourceforge.net/tracker/?group_id=77506&atid=550441>
> Would it be ok to ask for help with set up in this mailing list?
yes! we are all here to help. There is plenty of expertise on this list
and the way you are asking I am confident somebody with the right skills
and some time at hand will jump in and help you. I'm sorry I don't know
much about setting up the develeopment environment on OSX.
> UofT
acronyms, acronyms, can you expand?
> studying AI.
I guess artificial intelligence?
> 1) does it matter which platform I am developing on?
> (at school I program on Mac and Linux only)
no. as long as you get the code to build on your platform and provide a
simple patch to show proficiency of your tools, you're free to use any
platform you feel comfortable with. It will be a requirement for your
code to work on all platform when integrated in Hugin, but this is
something for which you'll get community help from people experienced in
those other platform. For example last year Tim developed Celeste on
Ubuntu. Harry helped him getting it to work on OS X and I helped him
getting it to work on Windows.
> 2) what level of detail do you expect to see in the proposal when it comes
> to implementation? (Obviously I will study the code as much as I can in the
> time left, but I cannot guarantee that I would be able to say precisely
> which data structures, functions etc. I plan to implement).
I don't think that we need so much implementation detail as to a
complete analysis of data structures and functions / API. What we want
to see is in the case of a project such as lens calibration (which
implies the development of algorithms, some of which are still
experimental) is:
* research of prior work. are there papers out there detailing math
formulas that may be interesting for the task at hand?
* a schedule with a break down of the different task that can be
planned. You can not predict that one algorithm will be useful as
intended, but you can predict the time it takes to test (in Octave or
similar tools), the time it takes to implement a stand-alone proof of
concept in C++ and the time it takes to integrate in Hugin.
* an objective, stating what the intended achievement is.
> Sorry for longish email and thank you for your time.
no worries, your email is very relevant and it is my pleasure to take
the necessary time to answer it.
Looking forward to read more from you. But first things first, get your
school assignment done :-)
Yuv
anand mishra
> anybody still around/interested other than León and Lukáš who are
> already set up?
>
> Yuv
>
> >
>
Yuval Levy
Galloth <lordg...@gmail.com> wrote:
> But I will not start programming until I will be selected, so I give
> up on hugin (which requires sending a patches) although I understand
> why you requiere it.
> But when you asked, I am interested. Is that topic still free? Should
> I write proposal even if I will work on hugin only after I got
> selected?
We're not looking for mercenaries here, we are looking for committers to
join our community which is all about giving and not about taking.
Selection for participation in Google Summer of Code is a privilege. We
are humbled by Google's generosity toward our project. We are equally
humbled by the generosity of all contributors to the community and by
the generosity of students who decide to apply.
The only feedback I can give specifically to the question you are asking
is that you are free to write a proposal and we will consider it like
every proposal.
Potentially we could even accept you and get you started during the
community bonding period. And we could decide toward the end of the
community bonding period to ditch you for whatever reason. No money for
you then. And let's not forget that we could fail you at mid term and at
the end of the program.
And you know what? I'm not interested playing this kind of games.
Because from your statements I get the impression that once Summer of
Code is over you will disappear, and all my efforts and Google's money
will have been wasted on paying for some code that we could probably get
done faster, better, with less effort and money.
My interest is to attract people to stick around in the long term, to
see them growing. It's an investment. Like any investment, initially it
costs more than it yields. Over time it bear fruits.
Your post motivated me to look back, to see how the return on our
investment was.
I am happy to see:
* Ippei (2007) still active on the Mac build despite having joined the
workforce.
* León (2007) still active on FreePV and applying to VLC this year to
add panorama playing capabilities there in a joint project that the two
organizations have tried to set up since last year.
* Zoran (2007) went on mentoring in 2008 and is still around despite
being very busy in his new job.
Out of five 2007 students that's not bad. Of the other two students one
was a superstar. Her code is great. She has moved on. The other one we
had to fail, but it was a good learning experience for both. He
completed his 2008 GSoC project with another organization successfully,
and we learn to introduce the patch requorement.
* Marko (2008) still around, integrated his code long after GSoC was over.
* James (2008) still contributing fixes. Integrated his code long after
GSoC was over.
* Tim (2008) ready to be a mentor or a student this year, depending what
occasion arises, and still contributing. Integrated his code long after
GSoC was over.
* Fahim (2008) still active on his branch with the masking code. It's
our shortcoming that his code is not integrated in Hugin yet because we
do not really know how to evolve the user interface to make use of the
advanced tool he developed.
* Onur (2008), who with Zoran has worked on one of the most demanding
projects of ours, and is still around but busy with his academic career.
Right now we'd need somebody to pull the strings together and bind his
project, with Zoran's 2007 project, into the Holy Grail of a patent-free
control points generator.
Out of five 2008 students. Well, there was a sixth, together with VLC,
but he desisted because he got a full time job. We hope to catch up on
that project with León who is applying to VLC this year.
I rather give one of the slots available to a deserving student - to
somebody like Lukáš who was too young for GSoC 2007 and became a
frequent contributor and committer without expecting us to be accepted
as a GSoC 2008 mentoring organization and without expecting to be
selected. Or to one of the students who have shown interest, passion,
curiosity and humility (Dev, Irena, Joe, Yulia, and please forgive me if
I forgot somebody of those who joined this mailing list in the last week).
And I rather give a slot back to Google for allocation to a more
deserving student at another project, than waste Google's money and the
time of my fellow mentors on somebody who does not seem mature enough in
my judgment.
Wow, that was a long email! I felt angry when I started the reply. I
feel better now. Should I hit the "send" button? I think so. I am still
inclined to black-list you, but I may be wrong. Maybe you will see what
in your attitude is incompatible with this community? Feel free to come
back any time you want, the door is not closed. In the meantime I share
this exchange from our mailing list, conveniently stored at
<http://groups.google.com/group/hugin-ptx/browse_thread/thread/e02061a1555edd56>
with the other mentoring organization. Just that they know who they are
dealing with in case you apply there too.
Yuv
Yuval Levy
> Yes I am highly interested
and how far have you advanced since the last time I gave you feedback?
<http://groups.google.com/group/hugin-ptx/msg/b198cd08b123ab04>
do you need any help? if you don't ask, we don't know how to help you.
Yuv
Irena Damsky
anybody still around/interested other than León and Lukáš who are
already set up?
just finished setting up the Debian on my computer, and hopefully, by the day after tomorrow (busy day tomorrow, probably won't have any time, after all, first work day and I had been sick home for the past week...) I'll be able to report back and let you guys know that I've managed to compile the project....
Irena
Yuval Levy
Galloth <lordg...@gmail.com> wrote:
> I think it is stab in the back, because
> I cannot explain myself to the other mentor.
A link to this post will be forwarded to the GSoC mentors list so that
interested GSoC mentors can read the explanations.
> I should have answer into hugin email list and than unsuscribe,
> but now, I am not going to subscribe back
Done it for you. The mail, in its entirety as below, has been forwarded
to hugin-ptx. The archives are world-readable, no need to be subscribed.
I have no further comment nor further wish to discuss with you.
Yuv
Galloth <lordg...@gmail.com> wrote:
> Hello Yuv
> First I should explain why I am writing to you and not to hugin email
> list. That is because I unsubscribe from that list. You have right
> that I am not good for hugin comunity. I am going to explain it later
> in the email. I also think that you should read it all, because of the
> reason decribed later.
>
> 2009/3/28, Yuval Levy <google@****.ch>:
>> Hello Jan,
>>
>> Galloth <lordg...@gmail.com> wrote:
>>> But I will not start programming until I will be selected, so I give
>>> up on hugin (which requires sending a patches) although I understand
>>> why you requiere it.
>>> But when you asked, I am interested. Is that topic still free? Should
>>> I write proposal even if I will work on hugin only after I got
>>> selected?
>
> Did my email offend you? It was not the purpose and I do not see why.
>> We're not looking for mercenaries here, we are looking for committers to
>> join our community which is all about giving and not about taking.
> I am not a mercenary and I am little offend by that. I am person with
> more than one interests and hobies so I want to choose wisely, what to
> do. The fact, that google offers money is motivating that is true. But
> I am willing to do only jobs, that interests me.The fact I am not
> wiling to work for huggin before being selected does not mean that I
> am not wiling to do anything. Mostly it is an answer to your request,
> that student should do unnecesary work. At last that is how I see it.
> You require that students to do patches and wrote:
>
> no worries about duplicate efforts.
>
> for you it is a bug hunt, for the students it is about showing us that
> they can hack the code base. There is no requirement for the student
> patch to be useful or to be applied to the code base when considering
> them for GSoC.
>
> So I as student should work why? Not for money, not to write useful
> code and therefore I do not want to write at all. When you asked about
> interested student, I replied because I was interested in the
> proposal, but I also told you, that I do not want to send these
> patches. Maybe I should write you why.
>> Selection for participation in Google Summer of Code is a privilege. We
>> are humbled by Google's generosity toward our project. We are equally
>> humbled by the generosity of all contributors to the community and by
>> the generosity of students who decide to apply.
>
> Yes GSoC is a priviledge, but it should be because only best people
> are accepted. Therefore I would understand that you want to saw their
> code, to choose the best. and therefore I asked, if you are interested
> in my proposal even without me sending the patches, because If I do
> not send the code, you can not compare me with the others.
>
> I do not agree with other parts of the paragraph, but it is my
> subjective opinion and probably the biggest problem at all.
>
> I see GSoC as simply agreement. Students offers his time and work to
> some organization. Orgaization offers know how,interesting problems
> and open source oportunities. Goggle simply offers some publicity and
> a money. I myself think that publicity is more important. I would
> never hear of most of these interesting organizations there without
> GSoC.
>
>
>> The only feedback I can give specifically to the question you are asking
>> is that you are free to write a proposal and we will consider it like
>> every proposal.
>
>> Potentially we could even accept you and get you started during the
>> community bonding period. And we could decide toward the end of the
>> community bonding period to ditch you for whatever reason. No money for
>> you then. And let's not forget that we could fail you at mid term and at
>> the end of the program.
>
> When I fail with my job, than I am not going to get money. I do not
> know, why you writing it. Are you suggesting, that you took student,
> take his work and let him fail? I hope not, but it is how it looks to
> me. When I am writing proposal I think if it is interesting and if i
> am able to and wiling to do it. I do not even think about time so
> much. And therefore not about money. What if I choose proposal that
> will take me 6 months to complete or more? In this perspective,
> googles money are not so much. But as I wrote earlier, it is a
> motivation to start, not pay.
>
>> And you know what? I'm not interested playing this kind of games.
>> Because from your statements I get the impression that once Summer of
>> Code is over you will disappear, and all my efforts and Google's money
>> will have been wasted on paying for some code that we could probably get
>> done faster, better, with less effort and money.
>
> You got a wrong impression. If the community is nice and the problems
> are interesting, I will continue. Are you interested in my opinion
> about you? I think not, so I am not going to write about it.
>
>> My interest is to attract people to stick around in the long term, to
>> see them growing. It's an investment. Like any investment, initially it
>> costs more than it yields. Over time it bear fruits.
>
> It is the same for the student. During the GSoC one can evaluate how
> the community is and if he wants to continue to work with it.
> Sometimes it can be done a lot more faster, like now.
> That is an impressive result. I am really fully surprised.
>> I rather give one of the slots available to a deserving student - to
>> somebody like Lukáš who was too young for GSoC 2007 and became a
>> frequent contributor and committer without expecting us to be accepted
>> as a GSoC 2008 mentoring organization and without expecting to be
>> selected. Or to one of the students who have shown interest, passion,
>> curiosity and humility (Dev, Irena, Joe, Yulia, and please forgive me if
>> I forgot somebody of those who joined this mailing list in the last week).
>
> It si your right to choose and I wish them good luck. But Why should I
> write proposal if i do not have a chance of being accepted?
>
>> And I rather give a slot back to Google for allocation to a more
>> deserving student at another project, than waste Google's money and the
>> time of my fellow mentors on somebody who does not seem mature enough in
>> my judgment.
> It is all same here, I am not going to give a slot, I am going to give
> my work to someone.
>> Wow, that was a long email! I felt angry when I started the reply. I
>> feel better now. Should I hit the "send" button? I think so. I am still
>> inclined to black-list you, but I may be wrong. Maybe you will see what
>> in your attitude is incompatible with this community? Feel free to come
>> back any time you want, the door is not closed. In the meantime I share
>> this exchange from our mailing list, conveniently stored at
>> <http://groups.google.com/group/hugin-ptx/browse_thread/thread/e02061a1555edd56>
>> with the other mentoring organization. Just that they know who they are
>> dealing with in case you apply there too.
>
> Well that is other part of it. I think it is stab in the back, because
> I cannot explain myself to the other mentor. I can explain them that
> my attitude could be different to their organization than to yours.
> (Right I should have answer into hugin email list and than unsuscribe,
> but now, I am not going to subscribe back do not worry). Because,
> hopefully, they would appriciate students work more than you.
> You are right I did not realized, that hugin email list is stored on
> the net. If I did, I would answer into the list, so my answer would be
> stored as well.
>
> Jan
>> Yuv
>>
>
> Jan Kastil
> gal...@jabbim.cz
Yexo
> anybody still around/interested other than León and Lukáš who are
> already set up?
>
> Yuv
discovered this software a few days ago but I've managed to compile
the project (the instructions on the wiki are very nice). Now I'm busy
creating panorama views from old holiday pictures, I'm starting to
like this software more and more :).
I'm interested in both the Processing of very large images project
(integrating in VIPS), but before writing a coherent application I
first want to find my way around the source code a bit better.
I read the other thread with some ideas for patches, and implemented
one of them, now where should I sent it? Should I send it to this
mailing list or attach it in the sourceforge bugtracker?
And another question: I'm I correct to assume this mailing list is the
primary means you use to communicate? I couldn't find any irc
channel / forum. There wiki mentions another mailing list on yahoo
groups, but that seems primary used by consumers, not the software
developers.
Thijs
Yuval Levy
Yexo wrote:
> I've managed to compile the project
on what system?
> I read the other thread with some ideas for patches, and implemented
> one of them, now where should I sent it?
email it here as an attachment, please.
> And another question: I'm I correct to assume this mailing list is the
> primary means you use to communicate?
correct.
I look forward to see your patch.
Yuv
Dale Beams
Dale
> Date: Mon, 30 Mar 2009 04:57:16 -0700
> Subject: [hugin-ptx] Re: Summer of Code Students
> From: t.mari...@gmail.com
> To: hugi...@googlegroups.com
Thijs Marinussen
>> I've managed to compile the project
>>
>
> on what system?
>
>> I read the other thread with some ideas for patches, and implemented
>> one of them, now where should I sent it?
>>
>
> email it here as an attachment, please.
>
reason". Since this is very small I'll make another one to demonstrate
my skills better.
Thijs
Tim Nugent
Some of you may know that I worked on Hugin/Panotools for last year's GSoC and developed the cloud detection code library 'Celeste':
http://ultrawide.wordpress.com/2008/11/12/using-celeste/
Some background:
-I code on Ubuntu and Centos Linux (and run XP and Fedora under Vmware). I also have access to a large Linux cluster here at University College London where I'm in the 3rd year of my PhD in Bioinformatics.
-I'm a keen photographer: below water I shoot with a Canon G7 in Ikelite housing with Inon wide angle and macros lenses; above water I use a Nikon D90 with Tokina 10-17mm fisheye.
-As stated above, I worked on Hugin last year; I also write Perl code for the BioPerl project.
I have checked with Yuv and Leslie Hawthorwn and have been given the OK to reapply this year. The two projects I'm considering are outlined below. Please let me have your comments and suggestions.
Best wishes,
Tim
Straight-line detection for automated lens calibration
-Either create a new tab or modify CP editor tab.
-Semi-automatic: Draw box around line using wxWidgets.
-Fully-automatic: use whole image rather than providing selection.
-Use x,y coordinates of selection to take subsection of image using VIRGA.
-Smooth subsection using Gaussian convulation to highlight regions of the image with high first spatial derivatives [1].
-Run Canny edge detector [2] on resulting subsection.
-The Canny detector is regarded as state-of-the-art [3] and is also a VIGRA function [4] therefore no external dependencies.
-For fine tuning, scale and threhold paramaters can be user adjusted.
-Threshold resulting image until only 1 edge remains.
-This is potentially difficult and will depend on good (semi-automatic) selection of the line.
-Mark line on image in GUI.
-User decides whether detection is correct. If it is not, go back and adjust selection or paramaters.
-Move across or down resulting edge and gather x,y coordinates.
-Add these coordinates as straight line control points.
-Run optimiser and save lens calibration data.
[1] http://homepages.inf.ed.ac.uk/rbf/HIPR2/canny.htm
[2] http://en.wikipedia.org/wiki/Canny_edge_detector
[3] http://en.wikipedia.org/wiki/Edge_detection
[4] http://hci.iwr.uni-heidelberg.de/vigra/doc/vigra/group__EdgeDetection.html
Bracketing Panorama Model
=========================
-I have already written a patch [1] which deals with bracketed images, but there is a lot of room to extend this.
-Automatic detection of bracket images based on EXIF data.
-Tab for manual selection/grouping of bracketed images.
-Modify panorama model so that controls points are only generated once, then these CPs are applied to corresponding bracketed images.
-Only first set of images are aligned, then alignemnt is applied to corresponding bracketed sets.
-Options to stitch each bracketed set separately (submit to PTBatcher) or merge to HDR.
-Possible integration of pfstools tone map opperators, though this depends on library availability - development currently in progress [2].
[1] http://ultrawide.wordpress.com/2008/11/16/hacking-hugin-part-1/
[2] http://groups.google.com/group/pfstools/browse_thread/thread/4794cd064f5fd18e
Harry van der Wolf
Welcome Yulia,
congratulations. that's not the easiest platform to deal with.
Yulia wrote:
> spent quite a bit of time trying to compile the code on mac os.
you're encouraged to sign up for a wiki account and update them.
> Instructions
> in the most cases are straightforward and easy to follow (there were a
> couple commands that are outdated.
I'm not sure about the current status of SVN, and even less in relation
> Right now the issue that I'm having is that when I try
> to stitch images together, it throws an error.
to building on OSX. Maybe Harry can recommend an SVN revision that is
known to work?
Hi Yulia,
As Yuv already mentioned: welcome.
Please specify which error you get when running hugin and in which SVN version it was.
I very often compile svn versions of hugin and regularly build OSX bundles. I compile more often than releasing a bundle and in the recent svn versions (last 2-3 months) I did not find an svn not to work. Occasionally "something" doesn't work and needs some fixing.
Please note that hugin has partly functionality of it's own and partly relies on external tools. This means that, during this stitching, you could get an enblend segfault error (patch available by Lukas (Stativ), but not applied yet to enblend trunk). Please let us know where it crashes and in what svn if it is hugin.
Other than that, errors are the daily bread of development. You're
invited to check if there is a similar report in the bug tracker and add
details there, or start a new ticket there.
<http://sourceforge.net/tracker/?group_id=77506&atid=550441>
yes! we are all here to help. There is plenty of expertise on this list
> Would it be ok to ask for help with set up in this mailing list?
and the way you are asking I am confident somebody with the right skills
and some time at hand will jump in and help you. I'm sorry I don't know
much about setting up the develeopment environment on OSX.
Please ask. I will not use this thread to expand already on what to do. I only want to mention that there are 2 options to work on OSX:
- the standard cmake way
- the XCode project way via the XCode IDE.
The standard cmake way is the best way to go get familiar with compiling hugin. I assume you already read the <http://wiki.panotools.org/Hugin_Compiling_OSX> as you mentioned some outdated info there.
It is also the best way for the community as the cmake approach is "universal" (and I don't mean the OSX meaning of universal).
> 1) does it matter which platform I am developing on?
> (at school I program on Mac and Linux only)
As Yuv also answered this question: It doesn't matter. Like mentioned above: the cmake approach (on OSX) is the best approach in this case as you can test both on Linux and OSX at the same time.
Last year Onur also worked on MacOSX in the gsoc 2008 feature matching project.
Hoi,
Harry
Lukáš Jirkovský
> ===================================================================
> --- hugin_base/nona/Stitcher.cpp (revision 3767)
> +++ hugin_base/nona/Stitcher.cpp (working copy)
> @@ -156,7 +156,8 @@
> //bool color = (bands == 3 || bands == 4 && extraBands == 1);
>
> // check if all other images have the same type
> - for (imgNr = 1 ; imgNr < pano.getNrOfImages(); imgNr++) {
> + for (UIntSet::const_iterator it = usedImgs.begin(); it != usedImgs.end(); ++it) {
> + imgNr = *it;
> vigra::ImageImportInfo info2(pano.getImage(imgNr).getFilename().c_str());
> if ( pixelType != info2.getPixelType() ) {
> UTILS_THROW(std::runtime_error, "image " << pano.getImage(imgNr).getFilename()
>
>
Hi,
First I'd like to thank you for the patch.
But I've some notes about it(as I've been looking at this bug some
time ago). Unfortunately there is one culprit – removing this loop
also removes check if the images are the same pixel type (nona seems
not capable of remapping images of different pixel type, at least it
didn't worked for me correctly (and even if it worked, there could
probably be problem with enblend (I'll test this later))). The cleaner
way how to fix this problem would be to run this loop only when the
other images are present. And the cleanest way how to fix it would be
fixing nona for better support of images with different pixeltypes if
it really is necessary (I mean it could be only my problem that images
were not remapped as I' expected).
2009/3/30 Tim Nugent <timn...@gmail.com>:
I like the Bracketing Panorama Model most.
Anyway, good luck with applications to everyone!
Regards
Lukáš "stativ" Jirkovský
Mokhtar M. Khorshid
majors graduate who is just starting computer science master studies.
I am very interested in working on 3D space transformations. In
concrete terms I am proposing to allow Hugin to take the camera shifts
and other camera movements into account while doing its calculations.
Accuracy of the current code can be improved even further even for
cameras fixed on tripods given the camera's dimensions and distance
between the lens' center and the axis of the tripod, although this is
not the focus of my project.
Based on the amount of information available the actual
implementations details may vary, but in general my project will
involve mostly geometric transformations of points, working with
OpenGL, and some mathematical operations. I am capable of doing this
because I am a mathematician with good computer graphics background.
I have participated in last year's GSOC program, and build the
wxWallCtrl for wxWidgets. See it in action here:
http://thunder-power.com/wxWallCtrlSample.zip (this is my own binary
sample, not an official wxWidgets release).
My official GSOC proposal holds my details, but feel to pose any
questions to me here.
Faruq
Hi,
I'm Faruque, second year PhD student at U of Wales, Newport under
Robotic Intelligence Lab (my homepage at: http://ril.newport.ac.uk/sarker/index.php).
I'm interested to work in processing large images in Hugin. We have
got a 16 mega-pixel Prosilica GE4900C CCD camera for tracking about 30
mobile robots. It gives about 50MB image file per frame using a huge
memory. I'm running my tracking algorithm on Ubuntu 8.10 64bit and
got some nice results of tracking markers (http://ril.newport.ac.uk/
sarker/index.php?pid=21).
I'm compiling latest 3768 Hugin from SVN and fixed some small issues
related to library location.
Now I've got this error following a compilation of sources from svn:
root:hugin# make
[ 4%] Built target huginbasewx
[ 6%] Built target huginlevmar
[ 7%] Built target huginjhead
[ 30%] Built target huginbase
[ 36%] Built target celeste
[ 44%] Built target huginvigraimpex
[ 44%] Built target celeste_standalone
[ 45%] Built target celeste_train
[ 52%] Built target huginANN
[ 53%] Built target huginlensdb
[ 54%] Built target open_file
[ 54%] Built target align_image_stack
[ 55%] Built target autooptimiser
[ 55%] Built target fulla
[ 56%] Built target nona
[ 56%] Built target pto2mk
[ 57%] Built target tca_correct
[ 57%] Built target vig_optimize
[ 59%] Built target matchpoint
[ 61%] Built target hugin_hdrmerge
[ 62%] Built target hugin_stitch_project
[ 62%] Building CXX object src/hugin1/hugin/CMakeFiles/hugin.dir/
GLPreviewFrame.o
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:47:1:
warning: "DEBUG_HEADER" redefined
In file included from /home/newport-ril/tmp/hugin/src/hugin1/common/
utils.h:27,
from /home/newport-ril/tmp/hugin/src/hugin1/panoinc.h:
70,
from /home/newport-ril/tmp/hugin/src/hugin1/hugin/
GLPreviewFrame.cpp:36:
/home/newport-ril/tmp/hugin/src/hugin_base/hugin_utils/utils.h:62:1:
warning: this is the location of the previous definition
In file included from /home/newport-ril/tmp/hugin/src/hugin1/hugin/
GLPreviewFrame.cpp:44:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLViewer.h:41: error:
expected class-name before ‘{’ token
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLViewer.h:68: error: ISO
C++ forbids declaration of ‘wxGLContext’ with no type
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLViewer.h:68: error:
expected ‘;’ before ‘*’ token
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
constructor ‘GLPreviewFrame::GLPreviewFrame(wxFrame*, PT::Panorama&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:171:
error: ‘WX_GL_RGBA’ was not declared in this scope
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:171:
error: ‘WX_GL_DOUBLEBUFFER’ was not declared in this scope
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:178:
error: invalid conversion from ‘GLViewer*’ to ‘int’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:178:
error: initializing argument 1 of ‘wxSizerItem* wxSizer::Add(int,
int, int, int, int, wxObject*)’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnNumTransform
(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:726:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:732:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:738:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnTextCtrlChanged
(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:759:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:775:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnTrackChangeFOV
(wxScrollEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:851:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnCrop(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1019:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnDrag(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1033:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnIdentify(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1048:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::TurnOffTools
(std::set<PreviewTool*, std::less<PreviewTool*>,
std::allocator<PreviewTool*> >)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1061:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1067:
error: ‘class GLViewer’ has no member named ‘Refresh’
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:1073:
error: ‘class GLViewer’ has no member named ‘Refresh’
make[2]: *** [src/hugin1/hugin/CMakeFiles/hugin.dir/GLPreviewFrame.o]
Error 1
make[1]: *** [src/hugin1/hugin/CMakeFiles/hugin.dir/all] Error 2
make: *** [all] Error 2
root:hugin#
and if I add this header in GLView.h
#include <wx/gtk/glcanvas.h>
it shows this:
[ 62%] Building CXX object src/hugin1/hugin/CMakeFiles/hugin.dir/
GLPreviewFrame.o
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:47:1:
warning: "DEBUG_HEADER" redefined
In file included from /home/newport-ril/tmp/hugin/src/hugin1/common/
utils.h:27,
from /home/newport-ril/tmp/hugin/src/hugin1/panoinc.h:
70,
from /home/newport-ril/tmp/hugin/src/hugin1/hugin/
GLPreviewFrame.cpp:36:
/home/newport-ril/tmp/hugin/src/hugin_base/hugin_utils/utils.h:62:1:
warning: this is the location of the previous definition
In file included from /home/newport-ril/tmp/hugin/src/hugin1/hugin/
GLViewer.h:27,
from /home/newport-ril/tmp/hugin/src/hugin1/hugin/
GLPreviewFrame.cpp:44:
/usr/local/include/wx-2.8/wx/gtk/glcanvas.h:67: error:
‘wxGLCanvasName’ was not declared in this scope
/usr/local/include/wx-2.8/wx/gtk/glcanvas.h:73: error:
‘wxGLCanvasName’ was not declared in this scope
/usr/local/include/wx-2.8/wx/gtk/glcanvas.h:81: error:
‘wxGLCanvasName’ was not declared in this scope
/usr/local/include/wx-2.8/wx/gtk/glcanvas.h:89: error:
‘wxGLCanvasName’ was not declared in this scope
/usr/local/include/wx-2.8/wx/gtk/glcanvas.h:100: error:
‘wxGLCanvasName’ was not declared in this scope
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
constructor ‘GLPreviewFrame::GLPreviewFrame(wxFrame*, PT::Panorama&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:171:
error: ‘WX_GL_RGBA’ was not declared in this scope
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:171:
error: ‘WX_GL_DOUBLEBUFFER’ was not declared in this scope
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnNumTransform
(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:726:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:732:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:738:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp: In
member function ‘void GLPreviewFrame::OnTextCtrlChanged
(wxCommandEvent&)’:
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:759:
warning: format not a string literal and no format arguments
/home/newport-ril/tmp/hugin/src/hugin1/hugin/GLPreviewFrame.cpp:775:
warning: format not a string literal and no format arguments
make[2]: *** [src/hugin1/hugin/CMakeFiles/hugin.dir/GLPreviewFrame.o]
Error 1
make[1]: *** [src/hugin1/hugin/CMakeFiles/hugin.dir/all] Error 2
make: *** [all] Error 2
Any idea?
TIA
Faruque
kevindra
done ? If yes, Does it not have any scope of improvements?
Please suggest me that.
thank you
- kevindra
Yuval Levy
thank you for sharing your thoughts about the two projects, and for
stepping in. You're our wild card this year - you volunteered to be a
mentor and I appreciate your flexibility to volunteer to be a student
when I asked you last week. I was worried when I saw the quantity of
active students compared with the quantity of active students at the
same time last year.
Tim Nugent wrote:
> Straight-line detection for automated lens calibration
> ======================================================
...
> Bracketing Panorama Model
> =========================
...
IIRC you're the first student to show interest in the Bracketing
Panorama Model, which as described would help Hugin catch up on
commercial stitchers that have an easier user interface to handle
brackets and tonemapping.
A related challenge would be handling the seams of full spherical panos
(the 360° seam, the zenith and the nadir) during the blending phase and
during the tone-mapping phase. Currently, tonemapping a 360°x180° is a
painful manual procedure.
Straight-line detection sounds interesting too, and there are plenty of
students aiming at it.
Whatever you choose, I look forward for your application
.
Yuv
Yuval Levy
Thijs Marinussen wrote:
> Attached a fix for bug 2166837 "nona opens unused files for no apparent
> reason". Since this is very small I'll make another one to demonstrate
> my skills better.
thank you for taking the time to build Hugin and submit a first patch.
Lukáš has already given you a thorough peer-review. I look forward to
see your improvements.
That said, you have demonstrated what we need in a patch:
- you interact well with the community
- you have the toolset to build Hugin
- you started hacking and produced a patch
I recommend that you take care of your application, the deadline for it
is April 3. You can submit an improved patch even after April 3. We will
try to review/consider patches and improvement that arrive during the
ranking period, April 3 to April 20. There is no guarantee as the
workload may be high, but chances are that if you submit a patch before
April 10 it will get reviewed and accounted in your application review.
Welcome on board and good luck!
Yuv
Yuval Levy
Mokhtar M. Khorshid wrote:
> I have participated in last year's GSOC program, and build the
> wxWallCtrl for wxWidgets.
Thank you for your openness and straight-forwardness with which you
filled the application in the webapp.
Based on your code in the wxWidgets repository and on feedback from your
last year mentor, you do not need to provide a patch against Hugin to
qualify. There is enough listed at
<http://wx.ibaku.net/changelog/?d=MMK> that speaks for itself.
I've asked Julian if he would hire you again and the answer was yes.
I trust you can easily build Hugin and hack the code (wxWidgets is the
GUI toolkit). You are welcome to do the patch exercise for completeness,
but it is totally up to you.
> feel to pose any questions to me here.
It's a sensitive question. I'll ask it here because I think it is
relevant also for the other students (who may want to answer it for
themselves and maybe even to us). You may answer privately if you prefer.
What are the chances that you will stay with our community after GSoC
and continue to maintain your code and further contribute code to Hugin?
I've asked Julian whether you are still participating in the wxWidgets
community and whether you are still contributing to the code. You know
the answer.
We see GSoC as an entrance door, as the beginning of a longer
relationship. I have the impression that your view of GSoC is more of a
self-contained projects. Veni vidi vici. And I have no doubt you can and
will win. But I want this to be win-win for everybody.
Loyalty and relationships can not be enforced and very often they also
can not be predicted. How do I know that a student will stick around and
become a committer? I don't. I try to do my best at making them feel
welcome, and I look for evidence where available.
In your specific case, I appreciate your openness and honesty about your
lack of interest in photography. Add the evidence from wxWidgets and I
get the impression that you see a GSoC project with Hugin/Panotools as a
time-limited effort. Correct me if I am wrong.
There is nothing wrong with that when it is done openly, and your
openness, added with your impressive CV, speak for you.
If I am right about the differing views that we have of GSoC, we can
either seek a compromise in which both you and the community win. A
compromize that can reconcile our need for maintainability with your
need to move on after GSoC is over.
The project you are applying to is one of the longer terms commitments.
It's open ended and I would like to assign it to somebody who is likely
to stay around for longer. There are other projects that are more
self-contained and if you want we can work together toward an
application on one of those.
Alternatively, I can help you seek another mentoring organization for
which it is not a problem that the relationship is time-limited.
Whatever you choose, let me reassure you that you have my support. Your
CV is impressive. You have good references. Your approach and your
communication are open, genuine and make me want to help you achieve
your goals. With Hugin/Panotools in GSoC if there is overlap/agreement,
or in another context.
Contact me any time.
Yuv
1.418.948.3374
Yulia Kotseruba
Hi Harry,
>
> Please specify which error you get when running hugin and in which
> SVN version it was.
>
whenever I called enfuse/enblend or exiftool. But apparently all I had
to do was to just move tools to where Hugin was looking for them or
provide path in preferences (either way it worked).
> Please ask. I will not use this thread to expand already on what to
> do. I only want to mention that there are 2 options to work on OSX:
> - the standard cmake way
> - the XCode project way via the XCode IDE.
I compile everything using the cmake command with debug flag set. But
I am still not able to step through all files when running with gdb.
Is there anything else I need to do to be able to debug it on Mac?
As for compiling instructions, everything is straightforward, except
they don't mention installation of exiftools. Also for building
enblend using fink, the command for installing dependencies looks
like this:
fink install lcms lcms-shlibs boost1.33 libtiff libtiff-shlibs libxmi
libxmi-shlibs libjpeg libjpeg-shlibs libpng libpng-shlibs
and it didn't work for me. Instead of libpng i had to use libpng3. I
don't know if it is worth mentioning on the wiki, but I spent some
time reading fink manuals to resolve that.
Now that's done, I got my first panos and I'm trying to work on some
patches. The issue is that next week is the last week of classes and I
have everything due this Friday and next Monday. I'm still working on
my application and reviewing some relevant papers, so that would be
done by the deadline. Is it ok if I submit the patch maybe a day or
two after the deadline for application? I am very interested in this
project since it is something that goes well with what I study at
school and plan to continue doing after school ends. It is the timing
that kills me. Next week I am going to be practically free.
>
Please don't look at application that I submitted today, I'm going to
work on it tomorrow.
Thanks,
Yulia
Yuval Levy
> Is it ok if I submit the patch maybe a day or
> two after the deadline for application?
*yes*
> Please don't look at application that I submitted today, I'm going to
> work on it tomorrow.
don't worry, Hugin, like so many open source project, is work in
progress. Your work in progress application fits nicely in :)
Yuv
kevindra
I am Kevindra Pal Singh, pursuing B.Tech 6th semester , Information
Technology, in Indian Institute of Information Technology, Allahabad ,
India.
I am very much interested in the "Automatic feature matching for
panoramic image". The main reason behind this is that I have done a
project in Image Processing which is the "Image Mosaics Generator".
It generates the panorama using the multiple images of the same
environment. I have done the following tasks and made this task
accomplished:
1. First of all, remove noise from the image using gaussian filter.
2. Detection of the feature points in all the images i.e. corner
points.
3. Matching of those feature points using NCC ( Normalized Cross
Correlation ).
4. Rejecting the outliers ( using RANSAC )
5. Combine images to form the complete mosaic.
I have good knowledge about Image Processing and I know much about
feature matching area as I explored that area.
Also, it will be good to make this project efficient and robust as the
control points detector is not so robust.
Thank you,
Kevindra
sumit
I, Sumit Sinha, am a second year student pursuing a Bachelors degree
in Computer Science and Engineering from Indian Institute Of
Technology, India.
I am highly interested in applying for Hugin/Panotools. I have decent
programming skills and have also done a number of relevant courses
like Software Engineering, Computer Graphics etc.. I have done work on
letter recognition as part of project based on Image Processing. I
have also been involved in projects relating to Computer Vision, and I
have made a software for School Auditorium Management.
How do I proceed with the application process....Am I allowed to
submit a patch after 3rd april.
Thanks
-Sumit
Mokhtar M. Khorshid
Let me try to answer your question as best as I can.
My chances of "staying" are not very low. I still am subscribed to
wxWidget's groups (and even have enhancements that I still want to
make to wxWallCtrl), I still wander on Clickteam's forums every now
and then, etc... But, as a dedicated double majors student, I am
usually occupied during semesters and I am free during the summer and
winter periods. What this means is that even though I may be around
following news, answering questions, or making ideas, my actual code
contribution will be significantly less (if at all). I am now shifting
from undergraduate study to graduate study. I do not know yet, how
much the work load will be.
As I mentioned, I am mostly interested in the computer graphics aspect
of Hugin, not the photography part (although I did some shooting to
test Hugin, and was very impressed by the results that I am
considering using it for some special purpose of mine). So I believe
we have two issues: interest and availability.
Let me try to suggest an action plan. Firstly, as long as there is
work in 3D space transformations or OpenGL (or any other computer
graphics work) I will be interested in doing it. I do not know how
much of that kind is in demand, but I know I will be interested on
working on that if it is available. Secondly, my research focus in
graduate school will be computer graphics, so whenever I can find a
topic of overlap between my studies and hugin, I will be able to
devote more time to contributing as it will double as studying for me.
But of course, my participation level will be uncomparable with that
during GSoC. During the summer I can work for 8 or more hours per day.
I am very thorough with my work, so whenever I'm working on something
I will do everything possible to get it done right and in time. If
things go behind schedule, I will stay all day and night to catch up.
Naturally, I can't do this all year. During semesters this can go down
to 1 hour per day (unless things get very interesting or I find
research overlaps). I also support any code I have written for as long
as I can. So regardless of how busy I am, I will always help those
that are working with my code/design, and step in to fix any bugs
found in my code. I don't know how much time regular committers in
Hugin are expected to devote per day during their work/study periods,
though.
My objective for this summer is: Get some good (graphics programming)
experience by working on an interesting project, make a good
reputation by leaving a positive impression (so that my work can be
used as a reference), and, if possible, do this as a GSoC project.
Last year was my first experience with an open-source project, so I am
still relatively new to the world of open-source development and have
a lot to learn about on going participation. You were right in that I
do have the "well defined project" concept, basically because that's
how most projects I worked on were: I have a set of objectives to
finish before some deadline. This does not mean that I "have to" keep
that way of thinking about projects, or that I "want" to keep it that
way either, it's just what I'm used to.
Now that you have a clear vision of how/why/when I would be available
and what my objectives are, you should be able to guide me to what
would work best for everybody. I have no problem switching to another
project if you feel my availability as described above won't cut it,
as long as it is interesting. If you feel it is acceptable for this
project or have any suggestions please let me know.
Yuval Levy
and sorry if I did not answer your introduction post with the whole
batch yesterday. I burned the midnight oil on other things while waiting
for the latest Hugin to compile.
Faruq wrote:
> I'm interested to work in processing large images in Hugin. We have
> got a 16 mega-pixel Prosilica GE4900C CCD camera for tracking about 30
> mobile robots. It gives about 50MB image file per frame using a huge
> memory. I'm running my tracking algorithm on Ubuntu 8.10 64bit and
> got some nice results of tracking markers (http://ril.newport.ac.uk/
> sarker/index.php?pid=21).
> I'm compiling latest 3768 Hugin from SVN and fixed some small issues
> related to library location.
> Now I've got this error following a compilation of sources from svn:
see if there is something wrong with svn3768. It builds fine on Windows.
Unfortunately I did not have an Ubuntu box available to try.
Are your libraries standard Ubuntu ones, or do you build them yourself?
AFAIK there should not be issues with library location on standard Ubuntu.
If you are building the libs yourself, what versions? specifically
wxWidgets? You'll need 2.8 or higher. Also the wxGLCanvas is not
activated by default in Windows. I don't recall if this modification is
necessary when building on Ubuntu as well:
<http://wiki.panotools.org/Hugin_SDK_(MSVC_2008)#wxWidgets>
maybe somebody who is currently running/using Ubuntu can help you more.
In the meantime, start working in parallel on your application. You need
to apply before April 3 to be eligible. You can always refine your tools
and provide a patch a few days later, as we will rank the student
proposals until April 20, but don't delay too much, it's a first come
first serve queue.
Good Luck
Yuv
Good Luck
Yuv
Dale Beams
http://www.tatteredmoons.org/hugin_svn.html
Build completes fine.
Dale
> Date: Wed, 1 Apr 2009 08:34:29 -0400
> From: goo...@levy.ch
> To: hugi...@googlegroups.com
> Subject: [hugin-ptx] Re: Summer of Code Students
>
>
Yuval Levy
kevindra wrote:
> I am very much interested in the "Automatic feature matching for
> panoramic image". The main reason behind this is that I have done a
> project in Image Processing which is the "Image Mosaics Generator".
Bruno already pointed you in the direction to dig for the proposal.
Did you build Hugin on your computer? You need to submit a simple patch
against a recent Hugin SVN version to show that you have the tools and
basic skills that it takes to be successful on this project.
Because the application deadline is April 3, I recommend that you first
submit an application on Google's webapp and then work as fast as you
can on setting your tools and providing a simple patch. We will be
accepting patches after application deadline. Candidates will not be
accepted after application deadline if they have not applied on the
Google webapp.
Good Luck
Yuv
Yuval Levy
sumit wrote:
> How do I proceed with the application process....Am I allowed to
> submit a patch after 3rd april.
Yes, you can submit a patch after 3rd april, but please hurry.
Before 3rd april: get yourself an account on the Google webapp and
submit an application proposal to Hugin.
ASAP, but acceptable after 3rd april, probably for another ten days or
so: build Hugin on your computer and submit a simple patch against a
recent Hugin SVN version. All we want to see is that you have the tools
and basic skills to hack Hugin. The patch does not have to be big or
complex. Just changing the color of a button or a text is enough to qualify.
Good Luck
Yuv
Joe Templeman
I know you said it is possible to submit it a bit late but would it harm my application if I dont? I figured I could always throw together a PC from my spare parts if I absolutly have to...
I guess it makes for a stonger application to do a more complex/useful patch, but as I'm short on time (with exams a matter of weeks away), I would probably have to make it simple.
Thanks
Joe Templeman
--
Joe
Mokhtar M. Khorshid
build right away when I first got the code). I had to manually get
enblend and set its path for CMakeSetup, since it was not part of the
SDK. You might want to consider adding it to the SDK.
On Apr 1, 1:50 pm, "Mokhtar M. Khorshid" <MokhtarKhors...@gmail.com>
wrote:
Yuval Levy
Mokhtar M. Khorshid wrote:
> Let me try to answer your question as best as I can.
Your answers, your sense of purpose and your maturity impress me. I
believe we can work out something that will captivate you and be of
interest for all of us.
This year, one of our mentors is Sébastien Roy of Université de Montréal
<http://www.iro.umontreal.ca/~roys/en_index.shtml> . He is researching
fundamental problems of 3D computer vision
<http://vision3d.iro.umontreal.ca/> .
He publishes much of the software produced in his lab under an Open
Source license and we found many points in common between
Hugin/Panotools/Lighttwist.
There are no project ideas posted on
<http://wiki.panotools.org/SoC_2009_student_proposals> because we have
not had the time yet to work them out.
I0m meeting Sébastien and Vincent on Friday to prepare specific project
ideas that you (and other interested students) can pick up.
Stay tuned for that. The rules at Google are that if there is agreement
between the student and the mentoring organization, the scope of an
application can be changed, even that radically. You have an application
in the webapp, we have time up until around April 15 to work something out.
> I still am subscribed to
> wxWidget's groups (and even have enhancements that I still want to
> make to wxWallCtrl)
how far advanced is that enhancement? do you think you could provide it
as "qualifying patch"?
> as a dedicated double majors student, I am usually occupied
understandable.
> Let me try to suggest an action plan. Firstly, as long as there is
> work in 3D space transformations or OpenGL (or any other computer
> graphics work) I will be interested in doing it.
There is a lot of work in 3D space transformation that is waiting for
your talent.
I'll share with you a vision.
Imagine you're in a room. A classroom, though any concave volume will
do. Projectors (or wall integrated touch screen displays,
rear-projectors, whatever combination) are scattered all over the place,
to cover the concave surface completely, and they project a seamless
immersive environment.
To do this we need:
- warping and blending, similar to Hugin, although the process is
inverse since we are splitting into individual images from a single
hemispherical or equirectangular input, and there are further challenges
such as real-time.
- We'll need to adapt the geometry to the position of the projectors;
and we'll need to blend the intensity of the projections in overlap
areas. Some sort of calibration mechanism, like Hugin has geometric and
photometric adjustments.
Lighttwist <http://vision3d.iro.umontreal.ca/en/projects/lighttwist/>,
Sébastien Roy's project, currently does the blending horizontally on a
cylinder only, and uses a fisheye image from the center of the cylinder
to calibrate.
My vision is to break out of the constraints of the cylinder, so to make
a truly immersive experience possible; and to break out of the
constraints of a regular geometry such as a dome (planetarium) or a
cylindrical screen.
At some point there may be individual cameras associated with individual
projectors, and the calibration may be in real time, accounting for
moving objects and changing conditions. But for now, static, one-time
calibration of the system before projection as done in Lighttwist will do.
My background is in business and economics. Such a system would enable
humanity to put a planetarium in every classroom for less than 5000$.
Today the cost of a planetarium this size is about 500.000$ and it
requires a fixed installation. It will enable us to share a common room
even if we are time zones apart. Amd plenty of other applications.
Lot of work in 3D space and challenges for skilled mathematicians and
computer scientists like yourself.
And no, I won't ask this to be completed by one Google Summer of Code
project (although, isn't it April's Fool today?)
> my research focus in
> graduate school will be computer graphics, so whenever I can find a
> topic of overlap between my studies and hugin, I will be able to
> devote more time to contributing as it will double as studying for me.
that would be great.
> But of course, my participation level will be uncomparable with that
> during GSoC
understandable. GSoC is full time. Normal participation in Open Source
is spare time (which is even less than part time). GSoC is a unique
opportunity to work full time on a project. Beyond that, nobody expects
the same quantity and intensity. Rare are those moments in life when one
can devote himself to a hobby with such intensity.
> My objective for this summer is: Get some good (graphics programming)
> experience by working on an interesting project, make a good
> reputation by leaving a positive impression (so that my work can be
> used as a reference), and, if possible, do this as a GSoC project.
You're on track to achieve this.
> Now that you have a clear vision of how/why/when I would be available
> and what my objectives are, you should be able to guide me to what
> would work best for everybody.
I think you would be an enrichment to this community. It could be the
idea you originally applied to. It could be the ideas that will be
tabled (hopefully soon) with Sébastien Roy. At this stage we are guiding
each other toward knowing each other. Your reply gave me a much better
insight into your motives. We'll work something out for the common benefit.
Yuv
Yuval Levy
> Just an update. I have just built the Install project (which didn't
> build right away when I first got the code). I had to manually get
> enblend and set its path for CMakeSetup, since it was not part of the
> SDK. You might want to consider adding it to the SDK.
well done. yes it was part of the older SDK (which is no longer
compatible with the repository because of new dependencies introduced in
GSoC 2008).
Yuv
Yuval Levy
> good to hear I have a few days after that to do the patch etc.
just make sure you head out to the Google web app and put in an
application for Hugin, even if incomplete. It is not enough to simply
have an account in the Google web app, you absolutely need to have an
application, even if incomplete.
http://socghop.appspot.com/student_proposal/apply/google/gsoc2009/<your
LinkID>?organization=hugin
put in a reference to a page on our wiki and keep working the proposal
there.
> I know you said it is possible to submit it a bit late but would it harm my
> application if I dont?
* you must submit an application as described above before April 3. If
you don't you're out and there is no way we can help you. The
application can be incomplete.
You have time between April 3 and April 15 to complete the qualification
task (patch) and articulate your application on our wiki.
Good Luck
Yuv
Mokhtar M. Khorshid
> I0m meeting Sébastien and Vincent on Friday to prepare specific project
> ideas that you (and other interested students) can pick up.
>
> Stay tuned for that. The rules at Google are that if there is agreement
> between the student and the mentoring organization, the scope of an
> application can be changed, even that radically. You have an application
> in the webapp, we have time up until around April 15 to work something out.
>
> how far advanced is that enhancement? do you think you could provide it
> as "qualifying patch"?
possible fixes & minor modifications last year. The enhancements are
significant enough that I would have loved to work on it on this
year's GSoC as a project, although that did not seem possible after
discussing it with Julian. Mainly I wanted to support different
surfaces of rendering, and use threads for smoother loading.
> There is a lot of work in 3D space transformation that is waiting for
> your talent.
to something as complicated as this. But if I can contribute to this
in any way, I'd definitely be very glad.
> I think you would be an enrichment to this community. It could be the
> idea you originally applied to. It could be the ideas that will be
> tabled (hopefully soon) with Sébastien Roy. At this stage we are guiding
> each other toward knowing each other. Your reply gave me a much better
> insight into your motives. We'll work something out for the common benefit.
Mokhtar M. Khorshid
the text on the column headers of the image list tab consistent.
Below is the content of the patch.
Index: src/hugin1/hugin/ImagesList.cpp
===================================================================
--- src/hugin1/hugin/ImagesList.cpp (revision 3768)
+++ src/hugin1/hugin/ImagesList.cpp (working copy)
@@ -330,11 +330,11 @@
m_configClassName = wxT("/ImagesListImage");
InsertColumn( 1, _("Filename"), wxLIST_FORMAT_LEFT, 200 );
- InsertColumn( 2, _("width"), wxLIST_FORMAT_RIGHT, 60 );
- InsertColumn( 3, _("height"), wxLIST_FORMAT_RIGHT, 60 );
- InsertColumn( 4, _("yaw (y)"), wxLIST_FORMAT_RIGHT, 60 );
- InsertColumn( 5, _("pitch (p)"), wxLIST_FORMAT_RIGHT, 60 );
- InsertColumn( 6, _("roll (r)"), wxLIST_FORMAT_RIGHT, 60 );
+ InsertColumn( 2, _("Width"), wxLIST_FORMAT_RIGHT, 60 );
+ InsertColumn( 3, _("Height"), wxLIST_FORMAT_RIGHT, 60 );
+ InsertColumn( 4, _("Yaw (y)"), wxLIST_FORMAT_RIGHT, 60 );
+ InsertColumn( 5, _("Pitch (p)"), wxLIST_FORMAT_RIGHT, 60 );
+ InsertColumn( 6, _("Roll (r)"), wxLIST_FORMAT_RIGHT, 60 );
InsertColumn( 7, _("Anchor"), wxLIST_FORMAT_RIGHT, 60 );
InsertColumn( 8, _("# Ctrl Pnts"), wxLIST_FORMAT_RIGHT, 60);
On Apr 2, 11:39 am, "Mokhtar M. Khorshid" <MokhtarKhors...@gmail.com>
wrote:
Tim
Works fine on a clean 9.04 install on another computer; I think it may
be related to a source install of wxWidgets, but I've tried re-
installing from source and re-installing via apt but still no joy.
Tim
Tim
option. Thanks to James for the pointer. For some reason re-installing
the wxwidgets packages via apt didn't work.
Tim