[hugin-ptx] Re: autopano-sift-c failed with error code: 2

> > Hello Thomas > Tested autopano and it still works. Further I have a license of PTAssembler and use it too.That has it's own control point generator named ptapicker. On the website you can find a lot of info about various generating methods.Tested in hugin and it does a good job too. Same p

[hugin-ptx] Re: autopano-sift-c failed with error code: 2

Hi Henk, Am Montag, 17. Juni 2019 11:52:21 UTC+2 schrieb Henk Tijdink: > > When you take Autopano Sift-C from the default settings, then yoy can take > away Autopano (A. Jenny) from the default settings too. > I will remove all other cp generator settings (except cpfind and align_image_stack).

[hugin-ptx] Re: autopano-sift-c failed with error code: 2

Hello Thomas When you take Autopano Sift-C from the default settings, then yoy can take away Autopano (A. Jenny) from the default settings too. It doesn't work on windows 7 and higher. and it is only a 32 -bits program. The newest hugin is only 64 bits in Windows. You can only use it in a Virtua

[hugin-ptx] Re: autopano-sift-c failed with error code: 2

Am Dienstag, 11. Juni 2019 08:38:14 UTC+2 schrieb Caetano Veyssières: > > If it doesn't work anymore, why is it still in the dropdown menu ? > > For historic reasons. It dates before cpfind. You can remove it in the preferences. I will remove it from the default settings for the next release.

[hugin-ptx] Re: autopano-sift-c failed with error code: 2

If it doesn't work anymore, why is it still in the dropdown menu ? On Monday, June 10, 2019 at 7:25:18 PM UTC+2, T. Modes wrote: > > > > Am Montag, 10. Juni 2019 15:38:12 UTC+2 schrieb Caetano Veyssières: >> >> I downloaded autopano-sift-C-2.5.1.tar.gz >> > autopano-sift-c is deprecated and not m

[hugin-ptx] Re: autopano-sift-c failed with error code: 2

Am Montag, 10. Juni 2019 15:38:12 UTC+2 schrieb Caetano Veyssières: > > I downloaded autopano-sift-C-2.5.1.tar.gz > autopano-sift-c is deprecated and not maintained any more (in the meantime for more than 10 years, mainly because of patent and license issues). >From the release notes: It is st

[hugin-ptx] Re: Autopano-SIFT-c --refine option fails

i never got this refine in Autopano sift c to work. Don't use it. Use cp clean instead. On Saturday, August 18, 2012 5:17:11 PM UTC+2, Kunlun121 wrote: > > I noticed that the --refine option does work when I load only 7 images of > my panorama into Hugin. No problems in fact. But with the whole

[hugin-ptx] Re: Autopano-SIFT-c --refine option fails

I noticed that the --refine option does work when I load only 7 images of my panorama into Hugin. No problems in fact. But with the whole thing, it refuses. Can anyone help out? On Saturday, August 18, 2012 1:24:26 PM UTC+2, Kunlun121 wrote: > > When I try to add --refine to my Autopano-SIFT-C

Re: [hugin-ptx] Re: autopano-sift-c

That worked for me. Thanks! On Wednesday, November 30, 2011 12:21:58 PM UTC, nadv...@suse.cz wrote: > > Hi, > > I had the same problem, on intel I7, 6GB of RAM. IMHO it is a RAM issue. > I was able to fix it with --ncores 4 option, so it processes 4 images at > once. > > Vladimir > > > On Wednesd

[hugin-ptx] Re: autopano-sift-c

Using the --ncores 4 suggested, I was able to get past the error. -- 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 th

Re: [hugin-ptx] Re: autopano-sift-c

Yes, I agree with you Jeffery. If CPFIND has option to analyze smaller versions than working will be very fast. On Mon, Dec 5, 2011 at 8:28 PM, Jeffrey Martin <360cit...@gmail.com> wrote: > > I have asked before... I wish I could do something besides only ask :( > I REALLY wish cpfind had a

[hugin-ptx] Re: autopano-sift-c

I have asked before... I wish I could do something besides only ask :( I REALLY wish cpfind had a way to analyze smaller version of images (it uses 1/2 size but default but it's either this or fullsize - nothing else!) when you have 50 (or 500) source images, it would be extremely helpful to e.

[hugin-ptx] Re: autopano-sift-c

I have the same problem with CPFind, like George Mahoney who started a thread called "Hugin - No control points"... Here is what I previously wrote: I have the same issue here (except there's no error with the 1st images, only after 7 or 8). I'm also running Win7 64 (i3 2100, 4GB RAM). I've tried

Re: [hugin-ptx] Re: autopano-sift-c

Hi, I had the same problem, on intel I7, 6GB of RAM. IMHO it is a RAM issue. I was able to fix it with --ncores 4 option, so it processes 4 images at once. Vladimir On Wednesday, November 30, 2011, Adam Weld wrote: > So it seems that CPFind is getting errors after 7 (16mp) images? I don't > thi

Re: [hugin-ptx] Re: autopano-sift-c

So it seems that CPFind is getting errors after 7 (16mp) images? I don't think it is a RAM issue, I have 8GB. Also note that no control points are added. Settings for CPFind are default. -Adam On Tue, Nov 29, 2011 at 11:15 PM, Adam Weld wrote: > Oddly enough, I just tried two of the images and

Re: [hugin-ptx] Re: autopano-sift-c

Oddly enough, I just tried two of the images and it worked fine. In total the panorama has 75 images, when I load only a couple CPFind seems to work, otherwise I get this output: Finding control points... "C:/Program Files (x86)/Hugin/bin/icpfind" -o "C:/Users/Adam/AppData/Local/Temp/ha60F7.tmp" "

[hugin-ptx] Re: autopano-sift-c

On 30 Nov., 01:33, Adam Weld wrote: > CPFind worked for one panorama, that was taken with a 100mm lens. So, very > little distortion across the images. > But for a panorama I took at 24mm, it couldn't find even a single CP. > ... This sounds very odd to me. I (like many others) use cpfind on a r

Re: [hugin-ptx] Re: autopano-sift-c

CPFind worked for one panorama, that was taken with a 100mm lens. So, very little distortion across the images. But for a panorama I took at 24mm, it couldn't find even a single CP. I just used Carlos' autopano-sift-c with the same set of images, and the outcome was 327 CPs. That is a huge differen

Re: [hugin-ptx] Re: autopano-sift-c

Well, nor autopano-sift-c work everytime. That's why I usually mark 2 or 3 control points manually between each image pair. Anyway, I have a version of autopano-sift-c that I use on XP and I can give you it. I'll try to send to your private email. Cheers, Carlos E G Carvalho (Cartola) http://cart

[hugin-ptx] Re: autopano-sift-c

My very good reason is that it never actually finds anything. :( On Nov 29, 3:06 am, kfj <_...@yahoo.com> wrote: > On 29 Nov., 05:49, Adam Weld wrote: > > > Hi, it looks like this file is needed for Hugin to automatically find > > control points. I am running Windows 7 x64. > > I cannot find a do

[hugin-ptx] Re: autopano-sift-c

On 29 Nov., 05:49, Adam Weld wrote: > Hi, it looks like this file is needed for Hugin to automatically find > control points. I am running Windows 7 x64. > I cannot find a download location for this file, I have looked through > many threads but the links are all broken. > The Hugin SourceForge

Re: [hugin-ptx] Re: autopano-sift-c working better than cpfind on small rectilinear images

On Sunday, 14 August 2011 at 15:11:17 -0400, Yuval Levy wrote: Sorry, just got round to reading this thread. > On August 14, 2011 05:47:43 am kfj wrote: > >> If, of course, we'd find that in certain use scenarios one definitely >> outshines the other, this would be helpful. Not that I expect this

Re: [hugin-ptx] Re: autopano-sift-c working better than cpfind on small rectilinear images

On August 14, 2011 05:47:43 am kfj wrote: > it's the image's content that lends itself better to be detected > with either one or the other detector. Most likely. > If, of course, we'd find that in certain use scenarios one definitely > outshines the other, this would be helpful. Not that I exp

[hugin-ptx] Re: autopano-sift-c working better than cpfind on small rectilinear images

On 13 Aug., 22:50, Yuval Levy wrote: > On August 13, 2011 03:54:58 am kfj wrote: > > My approach > > is to use cpfind by default (which most of the times does a very good > > job indeed) and only use apsc if cpfind fails - as a fallback option. > > This is the *blind* approach.  If we develop a s

Re: [hugin-ptx] Re: autopano-sift-c working better than cpfind on small rectilinear images

On August 13, 2011 03:54:58 am kfj wrote: > On 12 Aug., 16:14, Jeffrey Martin <360cit...@gmail.com> wrote: > > i've noticed in a few cases that autopano-sift-c does a better job > > finding CP's than cpfind. > > > > do i need to upload some example images? has anyone else noticed this > > issue? >

[hugin-ptx] Re: autopano-sift-c working better than cpfind on small rectilinear images

On 12 Aug., 16:14, Jeffrey Martin <360cit...@gmail.com> wrote: > i've noticed in a few cases that autopano-sift-c does a better job finding > CP's than cpfind. > > do i need to upload some example images? has anyone else noticed this issue? I'd not call this an 'issue'. Cpfind is using a differe

Re: [hugin-ptx] Re: Autopano SIFT-C-2.5.1 downloadsite.

When I did research on copyright laws, (in reference to japanese animation and fansubs) I located some international agreements that would prohibit fansubs in the US even though copyright had not been applied for, as they had been copyrighted in their own country. Essentially fansubs are not legal

[hugin-ptx] Re: Autopano SIFT-C-2.5.1 downloadsite.

On Oct 22, 9:13 pm, Henk Tijdink wrote: > Hello All > Probably no patent issues because it is french webspace, because the > EU has other rules then the USA. If someone can confirm that in UE the SIFT patent isn't applicable I can put the binary on my webspace ( in Italy) Cheers -- You receiv

[hugin-ptx] Re: Autopano SIFT-C-2.5.1 downloadsite.

On Oct 22, 3:13 pm, Henk Tijdink wrote: > Probably it is a straight download without login. Perhaps the builder > of the windows installer can integrate it.(Matthew Petroff or the > Panz) Unfortunately, this site has a landing page as well and cannot be linked to from the installer. Matthew --

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 15.10.2010 17:56, kfj wrote: Reverting to full frame fisheye and reoptimizing v+a,b,c created fine output. If you just use 6 around and sky, you cannot expect to generate reliable a, b and c data. I calibrated with 3+6+3 shots. Works fine. Bernd -- Bernd Hohmann DV Sachverständiger &

[hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 13 Okt., 18:36, Bernd Hohmann wrote: > On 13.10.2010 11:54, kfj wrote: > Reverting to full frame fisheye and reoptimizing v+a,b,c created fine > output. > If you just use 6 around and sky, you cannot expect to generate reliable a, b and c data. They will depend on your (insufficiently distri

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 13.10.2010 23:49, Bruno Postle wrote: I hate bugtrackers. Will do my very best. We need to use the tracker because there is enough traffic on the mailing list that bug reports get easily lost. I hate bugtrackers because they are usually not very well maintained. Thats why I use a giant c

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On Wed 13-Oct-2010 at 23:13 +0200, Bernd Hohmann wrote: Where to send the error report? We have a bug tracker on sourceforge: http://sourceforge.net/tracker/?group_id=77506&atid=550441 I hate bugtrackers. Will do my very best. We need to use the tracker because there is enough traffic on

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 13.10.2010 22:45, Bruno Postle wrote: Ok. cpfind Pre-Release 2010.3.0.294c413a5883 crashes here (Ubuntu 8.10 with latest libtiff* from Debian testing, cpfind selfcompiled) with "Segmention Fault" when processing LZW compressed tif files. Is this only with LZW tif files? In this testcase:

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On Wed 13-Oct-2010 at 15:42 +0200, Bernd Hohmann wrote: Ok. cpfind Pre-Release 2010.3.0.294c413a5883 crashes here (Ubuntu 8.10 with latest libtiff* from Debian testing, cpfind selfcompiled) with "Segmention Fault" when processing LZW compressed tif files. Is this only with LZW tif files?

[hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 13.10.2010 11:54, kfj wrote: is there any switch to force autopano-sift-c to create a buch broader distriubution of the control points? Probably not, but why use autopano-sift-c? Because cpfind crashes on some images (see my last posting). I gave cpfind a try on a scenery with a sea, a

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 13.10.2010 01:27, Bruno Postle wrote: It isn't in the 2.0 release branch, neither I found the sources for it elsewhere. Can you help here? cpfind is in the HG 'tip', i.e. you need a recent 2010.3.0 snapshot. The 2010.2.0 release doesn't have it. Ok. cpfind Pre-Release 2010.3.0.294c413a58

[hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 11 Okt., 22:21, Bernd Hohmann wrote: > Hi, > > is there any switch to force autopano-sift-c to create a buch broader > distriubution of the control points? Probably not, but why use autopano-sift-c? I reckon I am using the same lens as you (mine is called Walimex pro fisheye, but I think it'

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On Tue 12-Oct-2010 at 19:59 +0200, Bernd Hohmann wrote: On 12.10.2010 18:07, Dale Beams wrote: Pablo's CPFind (or whatever it's currently called) works well. I've got mixed signals from devs. Not sure if it's in the current hugin branch or if it still needs to be built separately. It isn't

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 12.10.2010 18:07, Dale Beams wrote: Pablo's CPFind (or whatever it's currently called) works well. I've got mixed signals from devs. Not sure if it's in the current hugin branch or if it still needs to be built separately. It isn't in the 2.0 release branch, neither I found the sources fo

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

Dale Beams schrieb am 12.10.10 18:07: Pablo's CPFind (or whatever it's currently called) works well. I've got mixed signals from devs. Not sure if it's in the current hugin branch or if it still needs to be built separately. cpfind is already in Harry's 2010.3.0 build for Mac. If it's not alr

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

I have 8 image around and a fisheye for the Zenith. cpfind hardly finds any controlpoints at all, while autopano-sift-c finds lots between the 8 images and some to the fisheye. Suggestions? Jan On Tue, Oct 12, 2010 at 10:07 AM, Dale Beams wrote: > Pablo's CPFind (or whatever it's currently ca

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

Pablo's CPFind (or whatever it's currently called) works well. I've got mixed signals from devs. Not sure if it's in the current hugin branch or if it still needs to be built separately. If you want a Patent free ACPD I'd suggest using it, or building it from source. I've been using it exclusiv

Re: [hugin-ptx] Re: Autopano-sift-c: better distribution of control points

On 12.10.2010 17:48, Jeffrey Martin wrote: as far as i can tell, autopano-sift-c is too slow, too inaccurate, and no one wants to work on it because some barbarians decided it was a good idea to patent it. Beside the patent issue: if a new controlpoint generator would need 6 hours to create so

[hugin-ptx] Re: Autopano-sift-c: better distribution of control points

as far as i can tell, autopano-sift-c is too slow, too inaccurate, and no one wants to work on it because some barbarians decided it was a good idea to patent it. so my advice would be to wait for something better to come along :-) probably very soon :-) On Oct 11, 10:21 pm, Bernd Hohmann wr

[hugin-ptx] Re: Autopano-SIFT-C status / release

On Sat 15-Aug-2009 at 11:25 -0700, Yuv wrote: >Bruno Postle wrote: >> On Sat 15-Aug-2009 at 17:41 +0100, Bruno Postle wrote: >>> >>> I get a segfault with the current trunk: >>> >>> ~/src/autopano-sift-C/APSCpp/autopano-sift-c --projection 2,111 >>> 2.5.2.pto DSC_8858.JPG DSC_8859.JPG DSC_8860.JPG

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Jul 19, 9:08 am, Nicolas Pelletier wrote: > Thanks Tom, Thanks Scott. > The new version works flawlessly. It is running faster I think (no > smoothing?). There was another patch that got applied along with the memory-leak- fix. Performance of the first stage roughly doubles by inlining two ac

[hugin-ptx] Re: Autopano-sift-c memory leaks

Thanks Tom, Thanks Scott. The new version works flawlessly. It is running faster I think (no smoothing?). It is processing now the high rez stack I sent it! Thanks guys. nick On Sat, Jul 18, 2009 at 5:13 AM, Tom Sharpless wrote: > > I have just replaced the old copy of autopano-sift-c.exe for

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Sat 18-Jul-2009 at 08:02 -0700, Tom Sharpless wrote: > >Bruno did that some time back. But I was unaware of it, so the >defective version stayed on my ftp site until this morning. I'm sure >there are still bad copies out there "for download". So I do think an >officially sanctioned distribut

[hugin-ptx] Re: Autopano-sift-c memory leaks

Hi Yuv On Jul 18, 10:14 am, Yuv wrote: > Tom Sharpless wrote: > > Yuv, please consider what could be done to improve distribution. > > have you committed the patches to SVN? this would be IMO the most > effective step to improve distribution at the moment. Bruno did that some time back. But I

[hugin-ptx] Re: Autopano-sift-c memory leaks

"hfov considered dangerous" To push just a little bit further: what about ending our dependence on "hfov" altogether? I suspect Dersch chose this way of specifying the angular resolution because it is fairly easy to guess, and because back in the early '90s it was not easy to find out the sensor

[hugin-ptx] Re: Autopano-sift-c memory leaks

Tom Sharpless wrote: > Yuv, please consider what could be done to improve distribution. have you committed the patches to SVN? this would be IMO the most effective step to improve distribution at the moment. Yuv --~--~-~--~~~---~--~~ You received this message beca

[hugin-ptx] Re: Autopano-sift-c memory leaks

Hi Bart & Bruno Bruno is right. Apart from the alignment option (which doesn't work anyway, and will be removed from the next version) APSCpp has no dependence on image size or ordering, and if you use a .pto file that gives a projection and fov for each image, it will likely find control point

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Sat 18-Jul-2009 at 04:11 -0700, Bart van Andel wrote: >2. At the end of the info screen (seen when APSCpp is called from the >command line without any parameters), it displays this notice: > >"Notice: for the aligning to work, the input images shall be > 1. All of the same dimension and scale

[hugin-ptx] Re: Autopano-sift-c memory leaks

Just a small note: 1. APSCpp version still displays 2.5.0 without SVN number, shouldn't this be added? It makes distinguishing between different builds (e.g. erroneous or not) a lot easier, for one. 2. At the end of the info screen (seen when APSCpp is called from the command line without any pa

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Fri 17-Jul-2009 at 17:04 -0700, Tom Sharpless wrote: > >It ought not be so difficult getting a patched autopano-sift-c, as it >seems in fact to be. There hasn't been a release because nobody who was seeing the problem tested the current SVN, I did ask. >Bruno and Yuv, please consider what c

[hugin-ptx] Re: Autopano-sift-c memory leaks

Hi Ryan On Jul 12, 6:47 pm, "Ryan Sleevi" wrote: > > Tom, what in your mind represents an 'ideal' memory size for APSCpp? My > investigations into 64-bit compilation began first because APSCpp would > regularly hork if I fed in all of my images, capping out at 2GB and > eventually faulting. It w

[hugin-ptx] Re: Autopano-sift-c memory leaks

Once again I forgot that Google mangles things that look like e-mail addresses, to protect us, I suppose. So the password part of that URL got clipped in the post. It is TKSpwd1, put that between : and @ in the above URL. Or go to ftp://tksharpless.net, enter the password TKSpwd1, double click

[hugin-ptx] Re: Autopano-sift-c memory leaks

I have just replaced the old copy of autopano-sift-c.exe for Win32 on my ftp site with the patched one. You can get it with this URL: ftp://tksftp:tksp...@tksharpless.net/autopano-sift-c.exe This is the actual executable, not an installer. Apologies to anyone who downloaded the defective versi

[hugin-ptx] Re: Autopano-sift-c memory leaks

It ought not be so difficult getting a patched autopano-sift-c, as it seems in fact to be. It doesn't ship with Hugin because of patent worries, and there does not seem to be any well known source of up to date binaries. Bruno and Yuv, please consider what could be done to improve distribution

[hugin-ptx] Re: Autopano-sift-c memory leaks

Hey Scott YOU FOUND IT! Way to go!! For those who haven't read Scott's patch (I was one of those until just now), the problem is a call to ImageMap_GaussianConvolution() that I added, to smooth the image before downsizing (antialiasing). I got it wrong. In near total ignorance of how Nowozin h

[hugin-ptx] Re: Autopano-sift-c memory leaks

If my time estimate is good, RC4 has the leak, RC5 does not. I'll upgrade and see. Thanks, nick On Fri, Jul 17, 2009 at 3:06 AM, Scott wrote: > > On Jul 12, 1:13 pm, Tom Sharpless wrote: > > > There have been several reports of out-of-memoryerrors while running > > "autopano-sift-c". > > > I

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Jul 12, 1:13 pm, Tom Sharpless wrote: > There have been several reports of out-of-memoryerrors while running > "autopano-sift-c". > I know of no case in which such a failure has been replicated by a > second observer.  I have not yet been able to get APSCpp to exhaustmemoryon > either Windo

[hugin-ptx] Re: Autopano-sift-c memory leaks

ok, so there is something wrong. Your's did not just run because you had access to more memory... 600megs is far from it. Thanks, nick On Thu, Jul 16, 2009 at 3:22 AM, Seb Perez-D wrote: > > On Thu, Jul 16, 2009 at 03:10, Nicolas > Pelletier wrote: > > Just to confirm, you are running a 64 bit

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Thu, Jul 16, 2009 at 03:10, Nicolas Pelletier wrote: > Just to confirm, you are running a 64 bit build of APSC. Correct? > Do you know what was the memory level at which it peaked? Yes, I build APSC myself in a 64 bit setup, so I imagine it must be a 64 bit build. Memory consumption was rathe

[hugin-ptx] Re: Autopano-sift-c memory leaks

Hi Seb, Thanks for the info. Just to confirm, you are running a 64 bit build of APSC. Correct? Do you know what was the memory level at which it peaked? Thanks, nick On Wed, Jul 15, 2009 at 4:50 PM, Seb Perez-D wrote: > > On Wed, Jul 15, 2009 at 14:35, Nicolas > Pelletier wrote: > > 36 * 15

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Wed, Jul 15, 2009 at 14:35, Nicolas Pelletier wrote: > 36 * 15mpx images will blow up after 30-32 images, regardless of the value > of maxdim. I've used the default of 1600, and tried many values, even down > to 100. Still an out of memory. I've just tried with 64 jpeg images, each 10,000x5000

[hugin-ptx] Re: Autopano-sift-c memory leaks

Already did. This is one of the other details I found weird. 36 * 15mpx images will blow up after 30-32 images, regardless of the value of maxdim. I've used the default of 1600, and tried many values, even down to 100. Still an out of memory. nick Timothy Leary

[hugin-ptx] Re: Autopano-sift-c memory leaks

By default all the images are down-sized so that the largest dimension is 1600 pixels. There is a command line option called -maxdim which you can set this setting with. Try setting this to something like 1200 or 800. On Wed, Jul 15, 2009 at 7:54 AM, Nicolas Pelletier < nicolas.pellet...@gmail.c

[hugin-ptx] Re: Autopano-sift-c memory leaks

Thanks Andrew, I appreciate. In my case, the out of memory happens in the first phase, i.e. when searching for keypoints for each individual images. I did not write down the exact numbers, but it did "keep" much memory between images. (For example, before processing image X, process at 200 megs. W

[hugin-ptx] Re: Autopano-sift-c memory leaks

I found the largest leak. And now my build only leaks a few KB, however I think the main problem is the amount of memory the application uses Let me summarize the memory usage with 2 images. 1) During GenerateKeyspp an enormousness amount of memory is allocated for image processing. For a single

[hugin-ptx] Re: Autopano-sift-c memory leaks

Tried some more cases to try and find something weird. Unfortunately, nothing. I consistently run out of memory, whenever I take big images (threeshold seems to be 36 images of 10 mpix). Bigger pictures and out of memory. I have not tried more smaller pictures. Of the people who stitch many pic

[hugin-ptx] Re: Autopano-sift-c memory leaks

I have only found 1 small memory leak so far... 16bytes * number of images I believe. Here is the fix. Index: APSCpp/APSCpp.c === --- APSCpp/APSCpp.c (revision 4038) +++ APSCpp/APSCpp.c (working copy) @@ -415,6 +415,8 @@

[hugin-ptx] Re: Autopano-sift-c memory leaks

On Sun, Jul 12, 2009 at 20:13, Tom Sharpless wrote: > In short, the reported failures remain undiagnosed.  So I am appealing > to all interested parties to try to create a repeatable case where > APSCpp runs out of memory. I ran APSCpp on a set of 104 images, each 3872x2592 (10 Mpix), full-size w

[hugin-ptx] Re: Autopano-sift-c memory leaks

I reposted on the other thread, but I'll give a few other details here. I seem to get the leak problem consistently. 31 15mpix images seem to be the limit. I've done a few panorama with multiple exposure lately and just going around and over (i.e. not down\bottom\no tripod shots) is 16 pics. At fu

[hugin-ptx] Re: Autopano-sift-c memory leaks

> I know of no case in which such a failure has been replicated by a > second observer. I have not yet been able to get APSCpp to exhaust > memory on either Windows or Linux. And when I run it under a leak- > checking debugger (MSVC on Windows, valgrind on Linux) I see no sign > of memory leaks

[hugin-ptx] Re: autopano-sift-c and "foreign" project files

Thanks Thomas, will do --Tom On Jul 6, 1:36 am, "T. Modes" wrote: > Hallo Tom, > > > If nobody has an objection, I intend to patch APSC so that it can load > > PTAssembler and PTGui scripts too.  Please post any other suggestions > > for improvements or fixes; I might do those at the same time.

[hugin-ptx] Re: autopano-sift-c and "foreign" project files

Hallo Tom, > If nobody has an objection, I intend to patch APSC so that it can load > PTAssembler and PTGui scripts too. Please post any other suggestions > for improvements or fixes; I might do those at the same time. when you look on loading script files, have also a look on http://sourceforg

[hugin-ptx] Re: autopano-sift-c and "foreign" project files

Thanks, Harry The memory leak question I will definitely look into. As for the other, my feeling is that someone else should try to make the "bells and whistles" such as align work right. They were pretty broken when I started doing surgery on autopano-sift, and I didn't even try to keep them w

[hugin-ptx] Re: autopano-sift-c and "foreign" project files

Hi Tom, When you request for sugestions for fixes, there are two: hugin-bugs-2795133 autopano-sift-c has SEVERE memory leaks The other one is not a real bug in the tracker but one mentioned in the mail (I didn't c