Pages: [1]   Go Down

Author Topic: Using Tilt with a pano-head  (Read 5754 times)

erpman

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 61
Using Tilt with a pano-head
« on: September 11, 2011, 09:52:35 am »

I´m trying to figure out the pitfalls involved in using a Tilt-shift lens, like the canon ts-e 45mm, with tilt to get larger dof when stitching.

The idea is to make single row panoramas with tilt, but I´m also wondering about the possibility of doing for instance a double row panorama by shifting the lens up for the top row and then down for the bottom, keeping the tilt in the same place. Can it be done, anyone have experience with this?
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Using Tilt with a pano-head
« Reply #1 on: September 11, 2011, 06:30:36 pm »

I´m trying to figure out the pitfalls involved in using a Tilt-shift lens, like the canon ts-e 45mm, with tilt to get larger dof when stitching.

That should be no problem, provided that the stitching application allows to control an offset parameter. An appication like PTGUI (d and e parameters) can cope with it, Photoshop possibly may not.

Quote
The idea is to make single row panoramas with tilt, but I´m also wondering about the possibility of doing for instance a double row panorama by shifting the lens up for the top row and then down for the bottom, keeping the tilt in the same place. Can it be done, anyone have experience with this?

It's possible, with even more need for the stitcher's offset capability. Whether keeping the same tilt in multiple rows makes sense depends on the particular settings and scene layout.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

erpman

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 61
Re: Using Tilt with a pano-head
« Reply #2 on: September 12, 2011, 05:28:25 am »

Can you explain what an offset parameter is?
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Using Tilt with a pano-head
« Reply #3 on: September 12, 2011, 05:35:31 am »

I use both tilt lenses and pano heads, but I haven't actually tried it in combination.

It will probably work though, but you may need to do some tricks. If you change tilt, shift or even just were you focus, the angle of view and distortion of the lens will change slightly. If the pano stitcher assumes that the same lens with the same settings is used for all images it may get a problem when stitching, since it will try to make the same distortion correction on all images.

In Hugin (the pano stitcher I use), what you have to do is to group together all images taken with a specific setting and instruct the program to consider that as a separate lens. A two row panorama with a different setting per row would thus yield two different "lenses". I sometimes do two row panoramas with different focus distances per row, and then I need to do this for the stitching to succeed.
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Using Tilt with a pano-head
« Reply #4 on: September 12, 2011, 05:41:56 am »

Can you explain what an offset parameter is?

To be able to stitch together images they must be perfectly linear, without barrel/pincushion distortion etc. No real lens is without distortions, so what the stitcher does is to apply a distortion model with a number of parameters and make trial-error optimization to reverse the distortion to get "perfect" images that then can be fitted together.

"Offset" is a parameter in more advanced distortion models. The simplest models assume clean barrel/pincushion distortion originating from the exact center. Real lenses may be a bit more complex than that, especially if they are tilted. I'm not sure how well the distortion models available in the popular stitchers today can model tilted lenses, I'm no expert on what type of distortion a tilted lens introduce. I guess one have to try and see.
« Last Edit: September 12, 2011, 05:44:00 am by torger »
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Using Tilt with a pano-head
« Reply #5 on: September 12, 2011, 09:50:43 am »

Can you explain what an offset parameter is?

In addition to what 'torger' said, successful stitching depends on images that are geometrical distortion free. Even most regular lenses exhibit a certain degree of decentering, and in a T/S lens it is promoted to a feature. So, in order to remove symmetrical pincushion/barrel distortion, the images need to be corrected to images that are perfectly centered before they are corrected for symmetrical/geometrical distortion. That is what the offset parameters (historically parameters 'd' and '' e') are for. Software such as PTassembler, PTGUI, and Hugin, all support those parameters.

Even for non-T/S lenses it can improve the stitching quality if (after regular control point optimization) an intermediate run of only these 2 parameters is done. The reason why these parameters are not routinely included in an optimization run is that it increases the number of variables and thus the complexity and the need for more controlpoints in the overlap region. For lenses that have a serious amount of shift applied (and Tilt usually also introduces a small amount of decentering), the added complexity of the optimal solution will often make a regular stitch impossible, or suboptimal at best.

In my experience, it works best for shifted images when you make a note of the amount of shift that was used, and immediately set that as a starting position for the lens. If you e.g. used 5mm vertical shift on a 24mm high sensor, then you already know exactly how many pixels (usually minus) offset you need to use (5/24 x # of vertical pixels), and it is better to input that manually rather than frustrate the optimization routine by having to search for it. Small amounts of decentering (such as introduced by tilting) can usually be found by using the optimizer, but for larger amounts it helps to get the required amount in the ballpark which will allow the optimizer to converge to an optimum solution faster, instead of getting stuck in another suboptimal local minimum of the solution.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Scott O.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 315
    • Photography by Scott and Joyce
Re: Using Tilt with a pano-head
« Reply #6 on: September 15, 2011, 12:19:13 am »

Interesting question which I have never considered.  I have taken single row pans tilted down without issue.  If I were to try a 2 row pan (which I may now!) I would tilt the bottom row down and keep the upper row level with no tilt.  Seems like this should work.

fotometria gr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 568
    • www.fotometria.gr
Re: Using Tilt with a pano-head
« Reply #7 on: September 15, 2011, 04:33:52 pm »

Interesting question which I have never considered.  I have taken single row pans tilted down without issue.  If I were to try a 2 row pan (which I may now!) I would tilt the bottom row down and keep the upper row level with no tilt.  Seems like this should work.
Why would somebody need a 2 row panorama? The way I understand it, is that a panorama is all about width. If somebody would want more height, he may as well use an even wider lens and thus make horizontal stitching even simpler. Cheers Theodoros, www.fotometria.gr
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Using Tilt with a pano-head
« Reply #8 on: September 16, 2011, 03:41:43 am »

Why would somebody need a 2 row panorama? The way I understand it, is that a panorama is all about width. If somebody would want more height, he may as well use an even wider lens and thus make horizontal stitching even simpler. Cheers Theodoros, www.fotometria.gr

The reason for making 2 row panoramas is for resolution, some of us use panoramas as some sort of "poor man's medium format", that is stitch together a mosaic of tele photo images to get a higher resolution image to make huge prints which look good even at close inspection.

In landscape panoramas you can also often have different focus distances on the rows, far on the top row and near on the bottom row, and that way get a wider DoF. Another alternative to increase DoF is to use tilt of course as discussed above in this thread.
Logged

fotometria gr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 568
    • www.fotometria.gr
Re: Using Tilt with a pano-head
« Reply #9 on: September 16, 2011, 07:43:09 am »

The reason for making 2 row panoramas is for resolution, some of us use panoramas as some sort of "poor man's medium format", that is stitch together a mosaic of tele photo images to get a higher resolution image to make huge prints which look good even at close inspection.

In landscape panoramas you can also often have different focus distances on the rows, far on the top row and near on the bottom row, and that way get a wider DoF. Another alternative to increase DoF is to use tilt of course as discussed above in this thread.
OK, but if you use a wider lens: a)You will get the greater DOF anyway, b)you may use MF film and thus get even higher resolution! The file will be of 365mb for each of every 6x4.5 you shoot! Out of my experience, nothing can much my Roundshot 220vr on panoramic landscape and a S/H one would cost as much as (or less) a Nikkor 14-24, but alternatively I would use my Contax 645 (or even my F3) not with my imacon 528c but with film! In my 30 years in photography, it has been proven to me that its best not to look for "absolute solutions" but to "solve things absolutely". For example if I  was to buy a P180 with my contax I would have a superb 20mpx back for 98% of my photography that would cover low light as well and high resolution 80mpx to do copies of some paintings, instead I prefer to have a D700 or D3S and do even better low light photography (with faster lenses as well), a 528c to do my copies on even higher resolution with true color (which I can't have with any single shot back), some extra film backs for the rare case I will need extra high resolution, I keep my low iso 22mpx shooting in the studio and a roundshot for panoramas which will cope even with different light distribution on the panoramic scene AND be able to shoot a panorama when things move (the later 2 can't be done at all with your method). All this has cost me less than half the money than the fabulous P180 and at the same time I serve my self much better for all my needs. Regards, Theodoros www.fotometria.gr
« Last Edit: September 16, 2011, 07:48:21 am by fotometria gr »
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Using Tilt with a pano-head
« Reply #10 on: September 16, 2011, 08:35:18 am »

OK, but if you use a wider lens: a)You will get the greater DOF anyway, b)you may use MF film and thus get even higher resolution!

The DOF thing is kind of an illusion, the longer lens will resolve smaller details in the scene (since it looks only at a small part of the scene at a time) and thus you need the circle of confusion to be smaller in relation to the scene. But if one is satisfied with the same resolution as you will get with a wider lens you can let the circle of confusion be larger, and then DOF in the resulting image will be the same regardless of focal length. But of course it would not make any sense to make a two row panorama and the scale down the resolution to match a one row panorama with a wide angle lens.

Put it in other words - DOF is related to the resolution. If you have low resolution system, then DOF will be wide (since circle of confusion can be large without noticing), but if it is high, it will be narrow. Regardless of method used to increase resolution (larger formats, making mosaics with pano heads) it will present the same DOF problem. Of course, one can choose to have larger acceptance of things in the scene being slightly out of focus and enjoy the higher resolution only close to the focal plane, but only then one gets the extra resolution "for free".

For all high resolution images which need large DOF techniques like focus stacking and tilting become interesting. If one doesn't need them one either doesn't need large DOF (or make compromises, which often is ok of course) or does not make high resolution images.
« Last Edit: September 16, 2011, 08:37:22 am by torger »
Logged

erpman

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 61
Re: Using Tilt with a pano-head
« Reply #11 on: September 16, 2011, 10:45:42 am »

soberle: That actually seems like a good idea. But would we then change the focus distance for the top row, or keep focus in place? If you have the ability to test this and return with some experiences I would be eternally greatful!

Yes, resolution for fairly big prints ("poor man´s MF") is the issue here, so a wide angle will not do.

I don´t have much experience with tilt-shift and plan to rent one for a 3-week project. Hence my somewhat limited knowledge on the subject  ;)
Logged

fotometria gr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 568
    • www.fotometria.gr
Re: Using Tilt with a pano-head
« Reply #12 on: September 16, 2011, 01:12:00 pm »

The DOF thing is kind of an illusion, the longer lens will resolve smaller details in the scene (since it looks only at a small part of the scene at a time) and thus you need the circle of confusion to be smaller in relation to the scene. But if one is satisfied with the same resolution as you will get with a wider lens you can let the circle of confusion be larger, and then DOF in the resulting image will be the same regardless of focal length. But of course it would not make any sense to make a two row panorama and the scale down the resolution to match a one row panorama with a wide angle lens.

Put it in other words - DOF is related to the resolution. If you have low resolution system, then DOF will be wide (since circle of confusion can be large without noticing), but if it is high, it will be narrow. Regardless of method used to increase resolution (larger formats, making mosaics with pano heads) it will present the same DOF problem. Of course, one can choose to have larger acceptance of things in the scene being slightly out of focus and enjoy the higher resolution only close to the focal plane, but only then one gets the extra resolution "for free".

For all high resolution images which need large DOF techniques like focus stacking and tilting become interesting. If one doesn't need them one either doesn't need large DOF (or make compromises, which often is ok of course) or does not make high resolution images.
I guess it all depends on the size of the print you want to do, in landscape usually subjects are further away and thus what you call circle of confusion is a lesser problem or may be close to non exist. I will insist that buying inexpensive MF (Pentacon 6, kiev 88cm) and using film could be a very good idea, good inexpensive glass could also be the correct thing to do (flektogon 50?). Don't forget that tilting a lens does affect resolution. The idea behind photography is not to master complex unreliable techniques that are so time consuming that become useless, but to be able to master light instead and reflect the drama. Finally I believe that you won't find many people producing (and print) higher resolution images than me (I already said I mainly use a Roundshot 220vr for many of my panoramas), it's just that resolution is not a panacea, in fact it may be a secondary aspect in photography (DR is always more important) as long as you have enough of it. Regards Theodoros www.fotometria.gr
 P.S. Soon there will be a new section in my site called "panoramas" there will be hundrends of images there, mainly done with the Roundshot, but there will also be stitched DSLR or 35mm film and MF (both film and digital).
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: Using Tilt with a pano-head
« Reply #13 on: September 16, 2011, 03:21:13 pm »

OK, but if you use a wider lens: a)You will get the greater DOF anyway, b)you may use MF film and thus get even higher resolution! The file will be of 365mb for each of every 6x4.5 you shoot! Out of my experience, nothing can much my Roundshot 220vr on panoramic landscape and a S/H one would cost as much as (or less) a Nikkor 14-24, but alternatively I would use my Contax 645 (or even my F3) not with my imacon 528c but with film! In my 30 years in photography, it has been proven to me that its best not to look for "absolute solutions" but to "solve things absolutely". For example if I  was to buy a P180 with my contax I would have a superb 20mpx back for 98% of my photography that would cover low light as well and high resolution 80mpx to do copies of some paintings, instead I prefer to have a D700 or D3S and do even better low light photography (with faster lenses as well), a 528c to do my copies on even higher resolution with true color (which I can't have with any single shot back), some extra film backs for the rare case I will need extra high resolution, I keep my low iso 22mpx shooting in the studio and a roundshot for panoramas which will cope even with different light distribution on the panoramic scene AND be able to shoot a panorama when things move (the later 2 can't be done at all with your method). All this has cost me less than half the money than the fabulous P180 and at the same time I serve my self much better for all my needs. Regards, Theodoros www.fotometria.gr

As you say each to his own needs and  it depends on what physical size you intend to have the panorama viewed or printed at. Last year I shot a panoramic cityscape as a three row stitched panorama ( the camera was a PhaseOne body with a 39mp P45+ back) and another very accomplished panoramic photographer shot  the same view at the same time with a medium format Seitz Roundshot with ISO 100 film . The digital version had clearly superior resolution, dynamic range (details in very bright highlights and down into the deep shadows)  and color. He had his film drum scanned which yielded much better resolution than an IMacon can produce. However the film version obviously took less time to shoot and far less work after making the exposures. The iamge was shot so it could be printed at 8 feet tall by over 24 feet long.
Logged

fotometria gr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 568
    • www.fotometria.gr
Re: Using Tilt with a pano-head
« Reply #14 on: September 16, 2011, 06:32:33 pm »

As you say each to his own needs and  it depends on what physical size you intend to have the panorama viewed or printed at. Last year I shot a panoramic cityscape as a three row stitched panorama ( the camera was a PhaseOne body with a 39mp P45+ back) and another very accomplished panoramic photographer shot  the same view at the same time with a medium format Seitz Roundshot with ISO 100 film . The digital version had clearly superior resolution, dynamic range (details in very bright highlights and down into the deep shadows)  and color. He had his film drum scanned which yielded much better resolution than an IMacon can produce. However the film version obviously took less time to shoot and far less work after making the exposures. The iamge was shot so it could be printed at 8 feet tall by over 24 feet long.
1. The P45+ (I have a lot of experience with) is a resolution and DR champion that can compete any of the recent backs, 2.The difference can't be of the magnitude you quoted and I'm definite on this, I do know both systems well. If everything was done correctly, the difference wouldn't worth consideration and could go either side (not on DR though)!
 If you did have such a difference, consider the following: 1.A poor or defective lens was used on the Seitz 2.The Seitz was badly focused (that's easy on the 220VR) 3.The drum scanner was not of the quality you think (resolution is irrelevant) 4.The nodal point was not set correctly on the Seitz. 5.The film wasn't correctly developed 6.The film was slipping on the drum of the Roundshot (did you measure the length that was shot and compare it with the quoted on the camera's LCD?) 7.You compared both images on 100% while you should compare same size prints! 60x180cm would do. 8.You had a wrong exposure on the Seitz (did you use Auto mode? or was it manual? and if auto, did you apply exp. compensation? did you have dark areas on the scene?) 9.Did you have enough battery to withstand the instant "battery suction" of the roundshot? 10.....etc.
 When I focus my 220vr I always do it on a Nikon body (F3 with P screen), mark the lens carefully and then refocus on the 220vr with the loupe and compensate for my marking! For exposure I always do auto with exp. compensation!
 What struck me most on your quote was the worst DR of the Seitz!!!! THAT SIMPLY CAN'T BE DONE!!! Not if you use Auto mode correctly! There was definitely something wrong with your Roundshot, lens or method! Do it again ;) Cheers, Theodoros www.fotometria.gr
Logged

Ellis Vener

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2151
    • http://www.ellisvener.com
Re: Using Tilt with a pano-head
« Reply #15 on: September 17, 2011, 02:34:08 pm »

I am sure you know what you know, but that doesn't mean I am wrong. You simply were not there, and don't know the hows and whys of the subject matter we were photographing on that particular day. etc. It was not my Seitz Roundshot - which is indeed an excellent camera - but I know the photographer who owns it who and was shooting with it and that it was in excellent mechanical shape as it had recently returned from the factory where it was inspected and serviced. I was not responsible for the scans but seriously doubt he would have accepted them if they were technically flawed.

Please don't turn this is into one of those pointless stereotypical internet forum discussions where for party a to be right, party b must be wrong.
« Last Edit: September 18, 2011, 08:07:43 pm by Ellis Vener »
Logged

fotometria gr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 568
    • www.fotometria.gr
Re: Using Tilt with a pano-head
« Reply #16 on: September 17, 2011, 05:26:11 pm »

I am sure you know what you know, but that doesn't mean I am wrong. You simply were not there, and don't know the hows and whys of the subject matter we were photographing on that particular day. etc. It was not my Seitz Roundshot - which is indeed an excellent camera - but I know the photographer who owns it who and was shooting with it and that it was in excellent mechanical shape as i had recently returned fro mthe factory where it was inspected and serviced. I was not responsible for the scans but seriously doubt he would have accepted them if they were technically flawed.

Please don't turn this is into one of those pointless stereotypical internet forum discussions where for party a to be right, party b must be wrong.
That was never my intention Ellis! I was only quoting on my knowledge and not based on my suspicion like others do. Cheers, Theodoros www.fotometria.gr
Logged

Scott O.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 315
    • Photography by Scott and Joyce
Re: Using Tilt with a pano-head
« Reply #17 on: September 20, 2011, 11:26:14 am »

soberle: That actually seems like a good idea. But would we then change the focus distance for the top row, or keep focus in place? If you have the ability to test this and return with some experiences I would be grateful.

Leaving for Colorado tomorrow.  I'll see what I can come up with.
Pages: [1]   Go Up