Image-stitching and virtual tour solutions My account Updates
It is currently Sat Oct 25, 2014 11:14 pm

All times are UTC + 1 hour




Post new topic Reply to topic  [ 357 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15  Next
Author Message
 Post subject:
PostPosted: Tue Aug 06, 2013 11:11 am 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
Destiny wrote:
I really do not like the idea of going from RAW to jpg and then to TIF...

??????

What do you mean?

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Last edited by klausesser on Tue Aug 06, 2013 11:16 am, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 11:39 am 
Offline
Member

Joined: Mon Jul 19, 2010 8:53 pm
Posts: 2137
klausesser wrote:
Destiny wrote:
I really do not like the idea of going from RAW to jpg and then to TIF...

??????

What do you mean?

Klaus

Destiny thinks you sugested to first convert RAW to jpg and then these jpg's to TIF. This is not what you suggested.

What I understood from you is that when doing a bracket of 5 of more photos you can just as well shoot jpg. When shooting jpg you would process them and save them in tiff before stitching.

_________________
Regards, Hans Keesom
I stitch and render for other photographers. Price: RMS^3, no cure no pay. If you want to concentrate on your business let me do the stitching for you.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 12:15 pm 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
Yep.. I did actually..

Destiny...

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 12:39 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
Destiny wrote:
We are slowly getting our heads around the the HDR software in my VR Drive.. Its really quite awesome... We are still working out some issues but we now understand so much more with this software.. And, Urs at RoundShot told me its getting some great updates soon too.. I realise that there is a RoundShot plugin in APG to add the xml file, but unfortunately with 9 bracketed shots, with my Mac, there is no way I can use it... :((

Destiny..

You should ask the VRDrive folk to offer an option to record one co-ordinate datum per shooting position so that you could use the XML file even when pre-processing bracketed exposures using the HRD mode.

Alternatively such a XML conversion/translation could be added to the VRDrive Import wizard.

I believe Aaron has custom firmware in his Panoneed that records both multi- and single- shot data per shooting position XML files so he can choose to pre-process or not and still have an appropriate XML file available. Perhaps that could be added to the VRDrive too?

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 12:52 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
HansKeesom wrote:
klausesser wrote:
Destiny wrote:
I really do not like the idea of going from RAW to jpg and then to TIF...

??????

What do you mean?

Klaus

Destiny thinks you sugested to first convert RAW to jpg and then these jpg's to TIF. This is not what you suggested.

What I understood from you is that when doing a bracket of 5 of more photos you can just as well shoot jpg. When shooting jpg you would process them and save them in tiff before stitching.

I see.

I suggested:

1) shooting RAW - use a dedicated RAW processor for "developing" the RAWs to TIFFs.
2) put the TIFFs into a dedicated HDR-processor, process them to HDR and use tonemappig resp. tonecompressing.
3) save the processed images as TIFFs.
4) use them in APG.

Alternative:

1) shooting JPG - preferable with large amount of shots like 200shots x 5 steps for example.
2) i myself prefer to "clean" the JPGs in Lightroom first.
3) put the "cleaned" JPGs into a dedicated HDR-processor, process/map them and save them as TIFFs.
4) stitch the TIFFs.

As i found out several days ago it works very well to import bracketed sets into APG 3.0.7 and use fusion there.
It´s not HDR what you get then - but aside from the use for IBL you don´t need HDR at all and it´s almost comparable qualitywise
what you can achieve this way

My suggestion to "develop" RAWs or "cleaning" JPGs are related to eliminating CAs/fringes or doing a bit of careful
sharpening and maybe denoising in this stage.

When i shoot more than 200 positions using 7 and more steps bracketing @1EV or 1,5EV i shoot JPG. Using RAW in such cases
doesn´t have any advantage as i experienced by excessively testing.

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 1:05 pm 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
I totally agree with suggestion 1... Thats what I do...

and.... Where you found "several days ago it works very well to import bracketed sets into APG 3.0.7 and use fusion", I unfortunately found I have issues since I only have 4 GIG of RAM.. but at least we found the cause why it will not work the same as you.. Its apparent that features of APG are system dependent.. To use all of the features you need to run APG in 64 bit with possibly 8GIG, but 16GIG + better... I just tried again with my new captured set.. APG has been sitting there for over an hour with nothing running other than internet..

Destiny..

klausesser wrote:
HansKeesom wrote:
klausesser wrote:
??????

What do you mean?

Klaus

Destiny thinks you sugested to first convert RAW to jpg and then these jpg's to TIF. This is not what you suggested.

What I understood from you is that when doing a bracket of 5 of more photos you can just as well shoot jpg. When shooting jpg you would process them and save them in tiff before stitching.

I see.

I suggested:

1) shooting RAW - use a dedicated RAW processor for "developing" the RAWs to TIFFs.
2) put the TIFFs into a dedicated HDR-processor, process them to HDR and use tonemappig resp. tonecompressing.
3) save the processed images as TIFFs.
4) use them in APG.

Alternative:

1) shooting JPG - preferable with large amount of shots like 200shots x 5 steps for example.
2) i myself prefer to "clean" the JPGs in Lightroom first.
3) put the "cleaned" JPGs into a dedicated HDR-processor, process/map them and save them as TIFFs.
4) stitch the TIFFs.

As i found out several days ago it works very well to import bracketed sets into APG 3.0.7 and use fusion there.
It´s not HDR what you get then - but aside from the use for IBL you don´t need HDR at all and it´s almost comparable qualitywise
what you can achieve this way

My suggestion to "develop" RAWs or "cleaning" JPGs are related to eliminating CAs/fringes or doing a bit of careful
sharpening and maybe denoising in this stage.

When i shoot more than 200 positions using 7 and more steps bracketing @1EV or 1,5EV i shoot JPG. Using RAW in such cases
doesn´t have any advantage as i experienced by excessively testing.

Klaus

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 1:24 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
Destiny wrote:
Its apparent that features of APG are system dependent..

Yes - but that´s an obvious and also well known fact regarding ALL applications . . ;)

The more you expect an application to do the more ressources it needs.

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 1:36 pm 
Offline
Member

Joined: Mon Jul 19, 2010 8:53 pm
Posts: 2137
Destiny wrote:
I totally agree with suggestion 1... Thats what I do...
..........
APG has been sitting there for over an hour with nothing running other than internet..

Destiny..

Don't make things more difficult then needed, send them over!

_________________
Regards, Hans Keesom
I stitch and render for other photographers. Price: RMS^3, no cure no pay. If you want to concentrate on your business let me do the stitching for you.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 5:45 pm 
Offline
Member

Joined: Thu Jul 25, 2013 8:23 am
Posts: 174
klausesser wrote:
Destiny wrote:
Its apparent that features of APG are system dependent..

Yes - but that´s an obvious and also well known fact regarding ALL applications . . ;)

The more you expect an application to do the more ressources it needs.

Klaus

That depends.....I just crammed all functionality of my controler into 32 Kb....(Kb Klaus not Mb) by transfering static data from the SRAM to the Flash ram....in fact a lot more code but it uses a lot less memory.

Also it's good common practice to be able to use all functionality of the software at the recommended configuration. But having said that Klaus, I have 8 Gb of memory in my machine. I load 7 tiffs....of 12 Mpixel each using 114 Mbyte in memory each.....to a total memory usage of 5.36 Gb max....(I had a browser window open at the time as well). It used 100% processor load (a good thing since that is what they are there for) on a 8 core machine it will use 8 cores (unless you specify otherwise) also a good thing since it will take about half the time. So neither memory usage nor processor usage can be a problem and yet the problems remain.

Greets, Ed.



_________________
Olympus OM-D with HLD-6, Fuji GX680, Samyang 7.5, Olympus 9-18, Sigma 19, Panasonic 14-45, Nikon 50 1.4 on Novoflex with tripod mount, Nikon 80-200, Panasonic 45-200, Fujinon 135, Fujinon 80, Fujinon 65, Fujinon 50, Gitzo Gilux Reporter 2, Sirui Ballhead and Panosaurus 2.0 NNP adapter, Motorized Panohead of Canadian (ironware)/Dutch (electronics and software) construction.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Aug 06, 2013 11:54 pm 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
I dropboxed my latest images for those who would like to test stitch them using apg 3.... I cannot... not without a lot of editing...

My shooting pattern was auto suggested by my VR Drive, -45 +45, which seems to work really well with PTGiu, but not APG 3.

Destiny

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 12:46 am 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
Destiny wrote:
I dropboxed my latest images fornthose who would like to test stitch them using apg 3.... I cannot... not without a lot of editing...

Destiny

Here's my results.

Standard settings, no post-processing.

1. APG 3.0.7 Editor view - stitch looks fine.

2. APG 3.0.7 rendered image - stitch looks poor.

3. PTGui Pro 9.1.7 rendered image - stitch looks perfect.
.........

What's going on with APG 3.0.7?







_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 12:59 am 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
I get excited when I see the preview, but... cry when I see the rendered pano. I have set detect to high, which took a lot longer than processing using PTGui which was perfect. Well, it makes me feel not so dumb at not being able to work it out. You are using the same images which provides for the best test situation. I need the test the stiching using APG 2.6..

Destiny

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 1:10 am 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
What I do not understand, the RMS is good, the preview is awesome looking, yet the render is yuck... Something is happening during rendering....

Destiny

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 6:27 pm 
Offline
Member

Joined: Mon Jul 19, 2010 8:53 pm
Posts: 2137
Destiny wrote:
I dropboxed my latest images for those who would like to test stitch them using apg 3.... I cannot... not without a lot of editing...

My shooting pattern was auto suggested by my VR Drive, -45 +45, which seems to work really well with PTGiu, but not APG 3.

Destiny

Working on them already ;-)

_________________
Regards, Hans Keesom
I stitch and render for other photographers. Price: RMS^3, no cure no pay. If you want to concentrate on your business let me do the stitching for you.


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 7:07 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
HansKeesom wrote:
Destiny wrote:
I dropboxed my latest images for those who would like to test stitch them using apg 3.... I cannot... not without a lot of editing...

My shooting pattern was auto suggested by my VR Drive, -45 +45, which seems to work really well with PTGiu, but not APG 3.

Destiny

Working on them already ;-)

Alex offered an explanation for why we might be having problems with these Photomatix fused image sets, and a workaround:

http://www.kolor.com/forum/p120113-today-12-23-29#p120113

The workaround worked very well with one image set, but not so well with another:

http://www.kolor.com/forum/p120123-today-14-17-50#p120123

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Last edited by mediavets on Wed Aug 07, 2013 7:08 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 7:41 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
mediavets wrote:
HansKeesom wrote:
Destiny wrote:
I dropboxed my latest images for those who would like to test stitch them using apg 3.... I cannot... not without a lot of editing...

My shooting pattern was auto suggested by my VR Drive, -45 +45, which seems to work really well with PTGiu, but not APG 3.

Destiny

Working on them already ;-)

Alex offered an explanation for why we might be having problems with these Photomatix fused image sets, and a workaround:

http://www.kolor.com/forum/p120113-today-12-23-29#p120113

The workaround worked very well with one image set, but not so well with another:

http://www.kolor.com/forum/p120123-today-14-17-50#p120123

Unfortunately Destiny explicitly expelled me from trying her images - so i only can state related to my own images (5D2 and D800) and the images from collegues over and over again: there is no basic issue with Photomatix-processd images in APG. I do it serveral times a week for years now. And other photographers (!) which i well know do it too and never had or have issues stitching the processed images.

So i doubt very much there´s an issue in the applications . . .

Can´t be it works for most of the users flawlessly but for some few it doesn´t. Sorry for saying that.

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 8:02 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
klausesser wrote:
i only can state related to my own images (5D2 and D800) and the images from collegues over and over again: there is no basic issue with Photomatix-processd images in APG. I do it serveral times a week for years now. And other photographers (!) which i well know do it too and never had or have issues stitching the processed images.

So i doubt very much there´s an issue in the applications . . .

Can´t be it works for most of the users flawlessly but for some few it doesn´t. Sorry for saying that.

Klaus

Klaus,

Have you seen Alex' report of his investigation of one sample image set pre-proceessed using Photomatix for exposure fusion that was causing stitching problems with APG 3.0.7 (but not with APG 2.0.9 nor with PTGui Pro 9.1.7)?

Alex did find an issue relating to APG 3.0.7, in this instance; he offered a possible explanation, and proposed a workaround pending further investigation:

http://www.kolor.com/forum/p120113-today-12-23-29#p120113

His workaround worked well for one sample image set, but didn't work well for second sample image set:

http://www.kolor.com/forum/p120123-today-14-17-50#p120123

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Last edited by mediavets on Wed Aug 07, 2013 8:04 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 9:23 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
mediavets wrote:
Klaus,

Have you seen Alex' report of his investigation of one sample image set pre-proceessed using Photomatix for exposure fusion that was causing stitching problems with APG 3.0.7 (but not with APG 2.0.9 nor with PTGui Pro 9.1.7)?

Alex did find an issue relating to APG 3.0.7, in this instance; he offered a possible explanation, and proposed a workaround pending further investigation:

http://www.kolor.com/forum/p120113-today-12-23-29#p120113

His workaround worked well for one sample image set, but didn't work well for second sample image set:

http://www.kolor.com/forum/p120123-today-14-17-50#p120123

Hey Andrew!

Of course i saw it - i can read you know . . . ;):cool:

WHAT was it that "caused stitching problems with APG 3.0.7"??? What´s the reason?

Alex related the issue to might-be focal-length differences. That´s exactly what i suspected all the time: if you use "align" the images might get cropped on their borders - which means: relatively to their original size that might be interpreted as a change of the focal-length in the EXIFs - OR NOT! If NOT there´s a difference between the real size and the size written in the EXIFs.

As i said before: a time ago i used "align" in Photomatix - the result was that PTGui refused to take the images. It said that the images were of different size and it can´t use them.
I don´t know how this is today - but i learned NOT to use "align" in Photomatix for images to stitch!

And i NEVER have ANY issue since then stitching Photomatix-processed images in APG.

The reason i - and others - don´t have these issue might be also related to the fact we use lenses which do not provide EXIFs at all.
For that reason i - and others - set the preferences for APG to NOT use the EXIFs at all.

So the ESSENTIAL question is: WHAT EXACTLY are the SETTINGS in Photomatix which causes the issues?

I say that because like said i and other collegues with i talked about this "issues" NEVER use "align" in Photomatix - and we NEVER had issues stitching the processed images in APG.

Causing issues by "aligning" images with "crop borders" in Photomatix comes with no surprise to me at all: it changes the original size AND maybe the image´s ratio which maybe produces a differece between what´s written in the EXIFs and what´s physically real . . .

As a photographer who encountered very, very strange issues over 30 years of practising commercial photography i´m hardly to come amazed . . . :D:cool:

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 9:52 pm 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
Klaus... You keep mentioning "Aline" images as a cause.. Yet I have stated many times I do not do this.. I have shown a screen shot with this turn OFF in Photomatix, yet you suggest that this would be a cause if I used it.. Well, I don't use it and I have never used it.. PTGui will not accept the images.. How I am getting a focal length difference with the Prime Fisheye Lens cannot be right either.. It is something else.. I have tried to use AlexJ work round, but it did not work for me.. Even though, I had a RMS of 3.86 with Quick Optimise and a 3.66 with Full Optimise stating its was Very Good, however the final stitch was NOT Very Good...

I have also eliminated RAM as the cause, since I did a test on Hans APG 3 PC, and it was the same...

Destiny..

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:01 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
klausesser wrote:
mediavets wrote:
Klaus,

Have you seen Alex' report of his investigation of one sample image set pre-proceessed using Photomatix for exposure fusion that was causing stitching problems with APG 3.0.7 (but not with APG 2.0.9 nor with PTGui Pro 9.1.7)?

Alex did find an issue relating to APG 3.0.7, in this instance; he offered a possible explanation, and proposed a workaround pending further investigation:

http://www.kolor.com/forum/p120113-today-12-23-29#p120113

His workaround worked well for one sample image set, but didn't work well for second sample image set:

http://www.kolor.com/forum/p120123-today-14-17-50#p120123

Hey Andrew!

Of course i saw it - i can read you know . . . ;):cool:

WHAT was it that "caused stitching problems with APG 3.0.7"??? What´s the reason?

Alex related the issue to might-be focal-length differences. That´s exactly what i suspected all the time: if you use "align" the images might get cropped on their borders - which means: relatively to their original size that might be interpreted as a change of the focal-length in the EXIFs - OR NOT! If NOT there´s a difference between the real size and the size written in the EXIFs.

Klaus

It was nothing to do with align - align was not used when processing these images in Photomatix.

Alex offered a speculative explanation, relating to focal length, but said he would undertake further investigations because the behaviour exhibited by APG didn't seem to really correspond with that theory.

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Last edited by mediavets on Wed Aug 07, 2013 10:09 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:16 pm 
Offline
Member
User avatar

Joined: Mon Apr 26, 2010 11:55 pm
Posts: 4903
Location: Australia
I had the idea that since part of the EXIF data was missing from the Photomatix images, that APG could no longer fully recognise the fisheye lens... So, a question for you Klaus.. Do you achieve an equally good results using a Fisheye as with a normal prime lens...??

I then decided to change Fisheye settings to Strong.. Although it was better stitching is was still not perfect...But, it was better... I also tried to set focal length from 'auto' to 'Force different' ,and for lens distortion from 'auto' to 'Force identic'. It made it much worse...

Destiny..

Klaus.. If you can get your hans on a 10.5 Fisheye, can you please test that using Photomatix images...

_________________
Primary School Teacher - Hobby Virtual Tours and 3D/360 Object/Product Photographer - Our aim is to create a 3D/360 Virtual Tour Shop/Store with links to 3D/360 degree products. My gear, VR Drive II - D90 - Nikon 10.5 fisheye - Nikon D800 with Nikon-Nikkor 14-24 - NOVOFLEX Magic Balance - Acratech GP Ball-head - Nodal Ninja 4 + RD-16 Rotor - Manfrotto 190CXPRO4 Carbon Fiber Tripod :) :)


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:34 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
Destiny wrote:
I had the idea that since part of the EXIF data was missing from the Photomatix images, that APG could no longer fully recognise the fisheye lens... So, a question for you Klaus.. Do you achieve an equally good results using a Fisheye as with a normal prime lens...??

I think Alex' earlier comments about Photomatix processed image EXIFs is a 'red herring' - I asked him to identify what if anything was missing from the EXIFs that APG might need and he hasn't mentoned it again in his recent report.

APG recognises the lens as a fisheye with a 35mm equiv. focal length of 15mm on D800 in DX mode which shoulkd be good enough.

PTGui Pro is able to extract all it needs from the same EXIF. And APG 2.0.9 managed just fine too.

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:38 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
Destiny wrote:
Klaus... You keep mentioning "Aline" images as a cause.. Yet I have stated many times I do not do this..

I thought about that as a POSSIBLE reason. I didn´t realize you NEVER used it.
Somebody nevertheless mentioned to use it with handheld-shots. Don´t remember who it was - thought it was you, sorry.

Thinking about and analyzing possible reasons is the only thing which can take things further.

Destiny wrote:
I have shown a screen shot with this turn OFF in Photomatix, yet you suggest that this would be a cause if I used it.. Well, I don't use it and I have never used it..

That wasn´t definitely clear so far.

Destiny wrote:
PTGui will not accept the images..

What do you mean?? Do you mean that PTGui didn´t take your images? Which particular images?

Destiny wrote:
How I am getting a focal length difference with the Prime Fisheye Lens cannot be right either.. It is something else.. I have tried to use AlexJ work round, but it did not work for me..

Well - that tells us: what?
"It´s something else" . . . right, must be. But what? I doubt it´s something in the app - other users would have the same issues. Do they?
To my knowledge they usually do not.

I´m quite sure it would be a big deal in the Photomatix-community - is it? Not to my knowledge.

Destiny wrote:
Even though, I had a RMS of 3.86 with Quick Optimise and a 3.66 with Full Optimise stating its was Very Good, however the final stitch was NOT Very Good...

No surprise: 3.66 is far from being "very good". "Very good" is around RMS 1 or 1.5 - at least with higher resolution than an average fisheye pano.
Having RMS 1 or 1.2 i can´t find errors at all - having RMS2 i find errors on the spot.

So that´s a bit hard to estimate what RMS is "very good" and what not. It´s less critical using fisheyes - here you simply don´t realize erros
as fast as with a 35mm @100% for example.
But using a D800 with a 16mm FE viewed @100% RMS2 it might also show errors viewed @100%. Didn´t try it so far.

I strongly suggest to ckeck over again very carefully whether the crop-mode of the D800 combined with the 10,5mm FE might cause the bigest issue - which takes influence on Photomatix also.

I suggest to send the *RAW* files to Kolor as well as to Photomatix for examination. Shot in DX mode and in FX mode using different lenses.

Having RMS2 with 35mm you definitely have errors in the stitch which you need to correct in Photoshop.

That´s why i´m after as much precision as possible . . . it takes hours and hours in a 700mpx or 4GPx pano chasing stitch-erors with indoor-shots.

Destiny wrote:
I have also eliminated RAM as the cause, since I did a test on Hans APG 3 PC, and it was the same...

RAM decides whether you can do it or not depending on the number of shots/resolution. IF it runs after all it needs to be good.

Conclusion: i have not the remotest idea what might cause the issue - YOUR issue. I know for sure that neither i nor other photographers i know have this issue.

Pity i couldn´t examine your shots.

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:44 pm 
Offline
Member

Joined: Mon May 22, 2006 12:18 am
Posts: 7809
Location: Duesseldorf, Germany
Destiny wrote:
So, a question for you Klaus.. Do you achieve an equally good results using a Fisheye as with a normal prime lens...??.

Yes - definitely.

Destiny wrote:
Klaus.. If you can get your hands on a 10.5 Fisheye, can you please test that using Photomatix images..
.

I used the 10.5 many times with Photomatix-images with the 20D and twice as a test with 5D2. Never had any issue.
A friend of mine uses it on a 3Dx - as a circular FE (shaved of course). No issues also with Photomatix.

Klaus

_________________
Simplicity is the keynote of all true elegance. Coco Chanel


Top
 Profile  
 
 Post subject:
PostPosted: Wed Aug 07, 2013 10:49 pm 
Offline
Member

Joined: Wed Nov 14, 2007 2:12 pm
Posts: 14047
Location: Isleham, Cambridgeshire, UK.
klausesser wrote:
Destiny wrote:
PTGui will not accept the images..

What do you mean?? Do you mean that PTGui didn´t take your images? Which particular images?

No Klaus.

Destiny meant that if you use align you often end up with images of slightly different dimensions and then PTGui will not accept them.

The test image set works fine with PTGui Pro and I've shown screenshots from it.

klausesser wrote:
Destiny wrote:
Even though, I had a RMS of 3.86 with Quick Optimise and a 3.66 with Full Optimise stating its was Very Good, however the final stitch was NOT Very Good...

No surprise: 3.66 is far from being "very good". "Very good" is around RMS 1 or 1.5 - at least with higher resolution than an average fisheye pano.

Irrelevant I think - if you read Alex' report you'll find that he got a perfect stitch with an RMS of 5.2.

_________________
Andrew Stephens
Many different Nodal Ninja and Agnos pano heads. Merlin/Panogear mount with Papywizard on Nokia Internet tablets.
Nikon D5100 and D40, Sigma 8mm f3.5 FE, Nikon 10.5mm FE, 35mm, 50mm, 18-55mm, 70-210mm. Promote control.


Last edited by mediavets on Wed Aug 07, 2013 10:50 pm, edited 1 time in total.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 357 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15  Next

All times are UTC + 1 hour


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group