(If there is a good thread here about focus stacking and someone could provide the link to it that would be awesome.)
Am I correct in thinking (from my research) that it's not possible to do focus stacking when there's been any movement at all in the camera from shot to shot? Thanks.
You are 99.5% correct. BUT, it depends on what you are trying to stack and how much extra Depth of Field (DOF) you need as well as other circumstances. Most photogs will tell you that you need to run a software program to focus stack (photoshop or one of those fancy plugins) - but I'll tell you that you can do it manually if you have the patience (and it helps when you have a 27" HD monitor like me - vs. a tiny crappy laptop).
Here are a few examples of images I've manually focus stacked where the subject (or the camera) moved from shot to shot because either it was handheld or there was wind.
The spider web was by far the hardest (and I think my first attempt). I had hand held the shot and it my camera moved between the split second frames (and so did the subject due to wind) but only needed a small section of one frame to give me more DOF so I was able to line it up and stack it.
Sometimes things won't line up perfectly (ok, most times they won't), but you can use the free transform tool to increase or decrease the size of something to match, or you can do a little cloning if needed. When I stack dragonfly shots, sometimes I'll just grab the eye from another frame that is in focus and essentially cut and paste it - but I'll often times have to enlarge or reduce the size of it to make it match perfectly.
Hi Barbara,
I'm just putting my toe in the water as far as stacking goes, but check out Scott Smorra's image "Dog Mountain Sunrise" posted on 7/10/11 if you haven't already. It's currently on pg 6 of the landscape gallery. A 4 image stack taken in windy conditions. Looks pretty darn good to me. However I would assume the margin of error is less for closeups like the ones Luke provided as opposed to landscapes. Maybe not. Hopefully some with more experience than me will chime in. Good question though since logic would dictate things need to be motionless.....but when you look at Scott's image......
Motion of the camera between shots is not a big deal, really, that can be easily corrected for by alignment done in PS or blending software.
Motion of subjects within the frame makes creating a clean blend just about impossible. While you can certainly make a blend like that look just fine in a small format web presentation, there will be artifacts in large prints that are in most cases impossible to remove completely.
The bottom line is, you cannot judge what a large print will look like in terms of detail and artifacts from a small format web image. If all you want is a small web image, then you can do all sorts of processing tricks
Posted by Garry Revesz on July 30, 2011 at 10:13 AM
I'm going to chime in here and ask the dumbest of dumb questions. I have heard of focus stacking but never attempted it (sounds very time consuming) but I can't help but ask; if you already have your shot lined up, adjustments made for the light, DOF set, etc., what advantage is there to stacking identical photos on top of each other? The only advantage I can see is if your first shot is at 2.8, then another at f8 and another at f16, or something like that. But a camera can't bracket f-stops, so you'd have to do it manually on a tripod. i can't quite wrap my little mind around this one. Great topic though Barbara
Garry
Posted by Garry Revesz on July 30, 2011 at 10:15 AM
I'm going to chime in here and ask the dumbest of dumb questions. I have heard of focus stacking but never attempted it (sounds very time consuming) but I can't help but ask; if you already have your shot lined up, adjustments made for the light, DOF set, etc., what advantage is there to stacking identical photos on top of each other? The only advantage I can see is if your first shot is at 2.8, then another at f8 and another at f16, or something like that. But a camera can't bracket f-stops, so you'd have to do it manually on a tripod. i can't quite wrap my little mind around this one. Great topic though Barbara
Garry, the idea is to focus manually at different focal planes with the same focal opening, thus creating images that are identical EXCEPT with different focal planes. Hopefully the DOFs would overlap and the merge would show the "in focus" portions of each image.
Barbara, Floris is right on. Camera motion can be eliminated by alignment in the stack. Subject or background movement needs to be almost nil. And you can get away with a lot in a small web post, but not in a print of any substantial size.
Posted by Dan Borzynski on August 1, 2011 at 3:09 PM
Then there is always the creative ways of abusing focus stacking, as when you put together images that you know won't produce clean images. Combine with a mid-magnification of insects (say 1/2 life size), there are always the 9 legged, two mouthed Japanese beetles, or combining two very different images and seeing what shows up.
But all in all, the above discussion has been fairly accurate.
The more movement usually the more clean up unless it is things like water or sky to which the software does a pretty good job sometimes of choosing (yet you can do simple masking there for choosing what sky or water you need). I should have a DVD series out on the issue in a couple of months.
Garry, the issue is probably more related to shutter speed. Say the wind is blowing and the light is low. Perhaps f2.8 is used in order to get the fastest shutter speed. Trying to use f16 or f22 to get the whole subject in focus at once makes the shutter speed way to long for a sharp image due to the wind. So, you take multiple shots at f2.8 changing the focal point in each take. Then you stack them in order to get most everything in focus in the final image. I understand the principles, but have not done it other than in one print of a clock tower. I needed more light (at dusk) to get a good image of the tower but it blew out the lighted clock. I then reset the camera for the bright clock. Later I copy/pasted the clock in to the image with the blown out clock.
Not looking for a reply, but under stacking, I wonder how come the out of focus regions of different images don't soften the areas with sharp focus....just have not researched the answer to this.
Posted by David Bostedo on September 11, 2011 at 12:41 PM
...I wonder how come the out of focus regions of different images don't soften the areas with sharp focus....
I'll reply anyway Troup. It's because focus stacking doesn't blend the various images. It actually picks and chooses the in focus parts of each and masks out everything else. So at any given point, you aren't seeing a blending of the multiple images; Instead you're seeing whichever was the most in focus of the whole stack. It's actually similar to merging and masking a panoramic photo, which is why it's part of the same tool in PS.
Discussion
10 repliesAm I correct in thinking (from my research) that it's not possible to do focus stacking when there's been any movement at all in the camera from shot to shot? Thanks.
You are 99.5% correct. BUT, it depends on what you are trying to stack and how much extra Depth of Field (DOF) you need as well as other circumstances. Most photogs will tell you that you need to run a software program to focus stack (photoshop or one of those fancy plugins) - but I'll tell you that you can do it manually if you have the patience (and it helps when you have a 27" HD monitor like me - vs. a tiny crappy laptop).
Here are a few examples of images I've manually focus stacked where the subject (or the camera) moved from shot to shot because either it was handheld or there was wind.
hairstreak butterfly
flower bud
spider web
The spider web was by far the hardest (and I think my first attempt). I had hand held the shot and it my camera moved between the split second frames (and so did the subject due to wind) but only needed a small section of one frame to give me more DOF so I was able to line it up and stack it.
Sometimes things won't line up perfectly (ok, most times they won't), but you can use the free transform tool to increase or decrease the size of something to match, or you can do a little cloning if needed. When I stack dragonfly shots, sometimes I'll just grab the eye from another frame that is in focus and essentially cut and paste it - but I'll often times have to enlarge or reduce the size of it to make it match perfectly.
Hope that helps!
I'm just putting my toe in the water as far as stacking goes, but check out Scott Smorra's image "Dog Mountain Sunrise" posted on 7/10/11 if you haven't already. It's currently on pg 6 of the landscape gallery. A 4 image stack taken in windy conditions. Looks pretty darn good to me. However I would assume the margin of error is less for closeups like the ones Luke provided as opposed to landscapes. Maybe not. Hopefully some with more experience than me will chime in. Good question though since logic would dictate things need to be motionless.....but when you look at Scott's image......
Motion of subjects within the frame makes creating a clean blend just about impossible. While you can certainly make a blend like that look just fine in a small format web presentation, there will be artifacts in large prints that are in most cases impossible to remove completely.
The bottom line is, you cannot judge what a large print will look like in terms of detail and artifacts from a small format web image. If all you want is a small web image, then you can do all sorts of processing tricks
Garry
Garry
Barbara, Floris is right on. Camera motion can be eliminated by alignment in the stack. Subject or background movement needs to be almost nil. And you can get away with a lot in a small web post, but not in a print of any substantial size.
But all in all, the above discussion has been fairly accurate.
Have fun!
Not looking for a reply, but under stacking, I wonder how come the out of focus regions of different images don't soften the areas with sharp focus....just have not researched the answer to this.
Troup Nightingale
Southeastern Photography
http://www.southeasternphotography.com
I'll reply anyway Troup. It's because focus stacking doesn't blend the various images. It actually picks and chooses the in focus parts of each and masks out everything else. So at any given point, you aren't seeing a blending of the multiple images; Instead you're seeing whichever was the most in focus of the whole stack. It's actually similar to merging and masking a panoramic photo, which is why it's part of the same tool in PS.