The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

using tilt for focus stacking?

Wayne Fox

Workshop Member
I believe I’m finally starting to get an idea of how tilt works and using it sometimes. I have stumbled across a few compositions where tilt can’t quite manage because the wedge of acceptable focus doesn’t cover the required elements in the image. Certainly I could use traditional focus stacking, but wondering if it is possible and indeed might be easier to use multiple shots where I move the wedge of acceptable focus instead.

I’ve attached a slice of an image were I struggled with this problem, the plant in the foreground is too tall to fit within the wedge so the top is out of focus. Additionally because the plant is on a hill the ground leading from the plant to the butte is actually going downhill so falls below the wedge. Also the top of the butte is slightly softer than most of the butte. so the question, can I tweak the wedge with focus/tilt changes then merge the resulting files like a focus stack? Or does changing the tilt make it difficult to merge the files. Seems 3 shots varying the tilt would get everything sharp, a focus stack would probably take 5 or 6.

Note: had trouble getting this to display at 1:1 on the web, so I took a 8800 pixel vertical slice, divided it in to three equal pieces and then lined them up side by side. Also shows better on web if I direct link it.

http://www.getdpi.com/gallery/files/4/2/2/8/gilsonbutte-tilt-problm_original.jpg
Any thoughts from those of you experienced using tilt to gain depth of field would be appreciated.
 

GrahamWelland

Subscriber & Workshop Member
Wayne,

I see no reason why you couldn't stitch multiple planes of focus taken with different focus wedges stacked vertically. You want the focus wedges to be taken for the highest vertical objects and then rotated downwards as needed until you've covered all planes of focus from top to the lowest image plane.

I'm sure that Helicon or manual blending of images in Photoshop would handle it just fine.

I've run into the same challenges with trying to get everything in focus with a single focus wedge. There's a trade off between maximum sharpness and having to stop down to f/16 or f/22 to create the widest wedge but then hitting diffraction. Better to stick to f/8 or f/11 and shoot a couple of wedges and blend I think.
 

studio347

New member
not sure if different tilting might make the covered(captured) areas slight different.. I mean front tilt.. Back tilting would make perspective change though. Brightness differences would be expected a bit, which would be not difficult to match through photoshop, I think.
 

GrahamWelland

Subscriber & Workshop Member
I would think that you'd probably only ever need a couple of shots to cover pretty much any situation so long as you're stopped down enough for a wide focus wedge. I agree though that there could be some perspective changes that would need to be considered.

Rather than totally reset the tilt, I'd take the approach of setting up the shot to optimize the shot with tilt as normal, ensuring in a landscape that I'd got the foreground and that plane in focus, perhaps at the expense of items at the top of the focus wedge (tree tops etc). Shoot and the tilt the camera up a degree or so and shoot again. You're only really interested in those elements outside the focus wedge.

It sounds like a fun technique to try. (I'm notorious for ending up with soft tree tops at times).
 

Shashin

Well-known member
Taking a variety of tilts and then stacking them would be an interesting idea. You would need the tilt to swing around the rear nodal point of the lens so the image does not move. The perspective should not change. But being lazy, I would ask why not just make a focus stack and forget the tilt altogether? Perhaps you would need only two images--the one with tilt and one with no tilt.
 

GrahamWelland

Subscriber & Workshop Member
:thumbup:

Any decent pano stitch or stacking app should handle any simple distortions even if you didn't tilt/rotate about the lens nodal point I would expect.
 

torger

Active member
I'm assuming that you would stack by changing focus distance, ie keep the tilt setting constant. This will keep the wedge axis fixed, and just turn the wedge around it as you change focus distance. If you change tilt then you would affect perspective a bit, even with rear tilt, which I would guess some focus stacking software would have problems dealing with. If you only turn the focusing ring I think you should only get normal focus breathing which focus stacking software is designed to handle.

In some situations you would get away with fewer shots using tilt-stacking than regular focus stacking, so it's surely an interesting idea. I haven't tried it myself though.

A problem you will have with the example scene you attached is that the wedge is very thin up close, so you won't get the full height of the flowers in focus if you want that. If you want that a flat focus stack would most certainly be a better option.

As a Linhof Techno user (digital field view camera) I use tilt a lot, even with wide angle lenses so I'm quite used to it. In the example you posted I would make a compromise, tilt about the same way, stop down a bit more perhaps and let some things be out of focus. I've made similar compositions and in many cases I like that you get slight differences in sharpness. In your example it's good to have flowers sharp, and top of the cliff sharp, but that the background behind the flowers is a bit less sharp I see as a plus, adds more depth to the image. I would stop down one more stop to make the effect a bit more subtle though, even if it would hurt resolution through diffraction a bit.

The attached image is an example of that type of composition. To get the top of the rock sharp I've aimed the wedge a bit high, and then the ground and lowest part of the forest behind the rock is a little soft (not visible on this scaled down image), while the forest at the top of the image has full sharpness. I think this contributes to depth, as you then get an edge between razor sharp rock and a bit softer background, but then just a little bit above the forest is razor sharp again. As long as the scene allows to keep these effects subtle I think it's a plus.

Looking at classic view camera technique one used to shoot with relative large aperture (f/64 on 8x10 corresponds to only f/9 on 135) and tilt to put the most important features into focus, and let the other be a bit blurry. Modern sharpening techniques in digital photograhpy lets us stop down more to keep the effect a bit more subtle.
 
Last edited:

Wayne Fox

Workshop Member
Thanks for the replies. Yes, my thought was to move the wedge by changing the focus distance, keeping everything else constant. So technique wise similar to a normal focus stack. My concern was whether the wedge could be moved far enough to get the entire flower (I’m only about 3 feet away from it). Seems I could get everything else with 3 positions pretty easily. Guess I’ll try it sometime.
 

Shashin

Well-known member
If you change tilt then you would affect perspective a bit, even with rear tilt...
You would want to only use lens tilt as that does not change perspective--rear tilt will. Perspective is controlled at the image/sensor plane.
 

torger

Active member
You would want to only use lens tilt as that does not change perspective--rear tilt will. Perspective is controlled at the image/sensor plane.
One can think of the lens as projecting a cone of light, and the angle of that cone will change when you tilt the lens, meaning that the sensor gets into another position in the cone. As the lens projects rectalinear perspective (ie stretch effect towards edges) any position change will affect perspective.

For front tilts and small angles it's usually negligible from a compositional point of view, but probably/maybe not from a focus stacking point of view. In focus stacking the software needs to compensate for focus breathing, very small changes in framing occuring from different focus distance. Tilt will cause asymmetric changes and I don't know if the focus stacking softwares can handle it.

You are right though that rear tilts were often used for perspective effects like looming, not sure thought it would have a larger effect than front tilt, wasn't it just that you usually used larger tilt angles at the rear? With front tilt and large tilts you need to compensate with shift though to not fall outside the image circle. If you do compensate with shift so the sensor becomes centered again it becomes equal as rear-tilt.
 

torger

Active member
Thanks for the replies. Yes, my thought was to move the wedge by changing the focus distance, keeping everything else constant. So technique wise similar to a normal focus stack. My concern was whether the wedge could be moved far enough to get the entire flower (I’m only about 3 feet away from it). Seems I could get everything else with 3 positions pretty easily. Guess I’ll try it sometime.
The wedge will be thin up close, and when not changing tilt (only focus distance) it will start from the same position so there may be difficulty to get enough "height" of the DoF up close. For example if you have high flowers close (say less than one meter away) and you want both the flower and the stem to be in sharp focus it may be difficult to achieve with tilt, probably also when tilt-stacking.
 

danlindberg

Well-known member
I’ve attached a slice of an image were I struggled with this problem, the plant in the foreground is too tall to fit within the wedge so the top is out of focus. Additionally because the plant is on a hill the ground leading from the plant to the butte is actually going downhill so falls below the wedge. Also the top of the butte is slightly softer than most of the butte. so the question, can I tweak the wedge with focus/tilt changes then merge the resulting files like a focus stack?
As you explain the scene, in my mind it is rather complicated introducing tilt. In this particular case I would keep the lens zeroed, stop down just before diffraction and focus stack.
I do use tilt often, but normally only to cover pretty flat ground from near to midview. Tilt becomes comlicated (imo) with big height differences in nearview.
 

stngoldberg

Well-known member
I have used tilt and focus stacking together many times primarily on landscapes (with a 120mm lens)where I want to achieve sharp focus on a foreground object. Helicon does a great job giving me what I want to achieve.
Stanley
 

cunim

Well-known member
I do this a lot with varying success. Here's an example in which I have a large vertical structure very close, and a long tubular structure extending off into the distance. Straight stacking is just not on.

Can't remember the tilt but, basically, it took in the top plane about half way into the distance. Then, stack something like 20 images using focus change. PITA to do but works OK if you can accept some softness at the extremes.

Peter
 

studio347

New member
sometimes, it feels very mechanical when I have to shoot 5 and more for focus. So, it might be fun to try more interactive ways to see if it works well... as mentioned, keep in mind the possible very subtle perspective_partial brightness variation( and the possible need for more than one LCC file, which is sometimes not so important.., I feel.).
Who knows? It might be much fun to do it~
For some personal reasons, I don't like the software method for focus stacking, which seems to be not perfect enough for me( at least, at pixel level..where the software is easily confused and makes a lot of compromise...it seems that I have nice outside costumes but not so good underwear...ha~). I think this is just a matter of personal taste...
 
Top