Profile cover photo
Profile photo
Daniele Sapino
109 followers
109 followers
About
Posts

Post has attachment
Hello +Torsten Simon​! Here is the comparison shot I promised.
I am posting one of 4 originals, a maximum of 4, and an average of 4.
I am sorry but I could not find a burst of more than 4 images to use.

However if you look at the waterfalls you can already see what I was talking about, the average composite is more silky, and would be generally more pleasing in general, but the maximum composite retains all the white drops and makes the falls a little beefier...

It is a matter of taste as well, but with few originals of this type I think maximum is a viable option too.
Photo
Photo
23/06/16
2 Photos - View album

Post has attachment
More about the stacking...

This is a stack of 7 originals using maximum mode.
I would say stacking is a decent alternative to long exposures with ND filters.

I used maximum instead of average because what I wanted was smoothing the white water flow on the dark background so, in this specific case, maximum did a slightly better job than Average, which was not bad either by the way.

The horrible edge softness that makes you feel dizzy has nothing to do with the stacking. It is the consequence of a cheap wideangle converter, hand held in front of the camera because I lost the needed step up ring.
As you can see the rock at the center is quite sharp, therefore the auto-align did its job nicely.
Photo

Post has attachment
Hello +Wasif Reza !

I was still thinking about the milky way stacking and i decided to make some tests...

I found a set of originals here:
http://photographingspace.com/homework-download-stack-data/

I think they are fit or the job even if a few caveats apply:
- they are all relatively bright already, so the thing to do is mostly noise control
- there is no foreground to take into account
- I assume they were taken with a motorized tripod because there is no star trailing across different photos.

I tried different stacking modes, and for each I am posting the stacked result straight out, and a version edited to match the exposure with an original which I intentionally pushed a bit too far from a single raw, without stacking.

I note that various modes yield different results in terms of white balance. I guess it is because of the change in chroma noise in the stacked image.

Add mode.
I combined just 2 images so that the exposure was not pushed too much.
Adding this way produces better noise than pumping a single raw, and by extension it should also be better than increasing iso in camera.
I guess however that without the above mentioned motorized head, star trails would appear instead of increased brightness.

Average mode.
By far my favorite, noise goes practically down to zero. For this mode I post a "to taste" version which is much darker than the others, because i consider the pitch dark sky between the stars very good looking. However, also in the "boosted" version, noise reappears, but I think it is still the best of the bunch.
Again, without a motorized tripod, star trailing would cause the stars to fade, and the presence of a foreground, with a moving tripod head may complicate things further.

Minimum mode.
I tried this under the assumption that stars are always bright, while noise comes and goes, so in theory minimum should mute the noise.
In reality this does not work as well as in theory. I guess it is because of the presence of a "baseline" noise and the flickering of the stars due to atmosphere.
Still, not really too worse than the rest, and much better than the pumped single raw anyways.
PhotoPhotoPhotoPhotoPhoto
2016-06-20
9 Photos - View album

Post has attachment
North Atlantic #puffins  at the tip of #latrabjarg  peninsula, in the #westfjords  of #Iceland .
Shot on Olympus XZ-2, edited to taste in PhotoMate R3
Photo

Post has attachment
Ihagee Auto Ultrix (bellows camera circa 1935)
Kodak Portra 800
Negative re-photographed and edited with PhotoMate R3.
Photo

Post has attachment
Here is another experiment I did, since snowflakes did not really cut it for trying "Median".

I used my table and some origami flowers to impersonate an architectural/street shot.

These pictures were taken with the camera leaning on the table, actuating the shutter by hand, therefore there was some offset between a shot and the other.
I left the auto align function on while stacking, in order to compensate for this. I don't know if this was a good idea or if it is the cause of the issues which I describe below.

Maybe +Torsten Simon  can shed some light about these observations, I am sure that at least some of the issues I had are because I am doing it wrong ^.^

- The first thing that I noted is that with the quantity of shoots involved in "clearing up" the shot, even in this "not so crowded" test, my tablet crashed.
I downsized the images and retried, and this time it worked like a charm.
I guess merging 10 pictures at full size is too much for the hardware. It is a pity because using too few images there increases the risk of not having all the needed spaces "empty", and resizing is always a bit of a bummer :(

- The second thing I see is that the program managed to remove most of the wandering "flowers" while the one that stays still in all but one photo remains visible. This is is exactly what I hoped for.

- I see however that a faint "ghost" of one of the removed flowers still appears to the right of the still one, and if you squint very hard, another less obvious ghost is there further to the left.
This is strange because for how I understod "median in statistics" either the flower is there, or it is not.

- I also see some vertical banding which the originals did not have.

- Finally, there are halos, most obvious around the kitchenware in front of the window, and around the still flower whihc also looks a bit jittery. As I said initially, maybe this is because of some glitch with the auto align feature which does not like moving objects.
PhotoPhotoPhotoPhotoPhoto
2016-02-29
11 Photos - View album

Post has attachment
Hello!

In the weekend we had some new photo stacking options... and quite bad weather (at least here in northern Italy), so I did only indoors testing, but I discovered that with some immagination snowflakes under a street lamp make a decent impersonation of what star trails would have looked like.
(I also discovered that the lamps in my street are not constant, just pulsing too fast for the naked eye to see).

In the gallery I propose you a series of four original images, and the stacked results in Minimum, Maximum and Median mode.

Here are some toughts, unfortunately without any technical background I can only go so far in understanding what goies on exactly:

- All images were taken handheld at 1/20 second, so a bit of auto alignment was required. Odd enough the auto alignment worked fine with the "Minimum" and "Median" elaborations, but failed in "Maximum", which is odd because I thought that the alignemnt was done first, and that the results would have been the same for all the modes.

- The results of the "Maximum" stacking are quite obvious, and in line with what I expected. I am only afraid that if I was doing star trails instead of snowflakes, the alignment glitch would have affected the trails as well.

- Ok, I felt dumb for writing the sentence above, star trails would obviously call for a tripod so little to no alignment would be required.
However, "Median" could be used to remove an unwanted passer by in an architectural shot or other situations where handheld would be an option.

- Since I was basically stacking white dots on a black background, I expected the result of "Median" and "Minimum" to be bascially the same, instead "Minimum" suppressed a lot more snowflakes than "Median".

What about you? Has anybody else experimented with these new modes? 
PhotoPhotoPhotoPhotoPhoto
2016-02-29
7 Photos - View album

Feature proposal - Focus stacking

Based on a set of pictures shot with focus bracketing, the software scans the same area of different pictures and chooses from each one the part in focus (i think it is done based on local contrast, but I am not sure). It is mostly useful to get macro shots with very deep depth of field.

Feature proposal - Dark and light field composite.

I found this in the "specials" section of the om-d e-m10 mk II (olympus feature-naming and camera-naming teams are both a bunch of sadists).

Basically it is a way to stack multiple photos (generally a burst of long-ish exposures) keeping for each pixel only the brightest value across the different images, so areas which do not change of illumination do not become overexposed, dark areas stay dark, but things like fireworks, lightnings, car streaks, light painting or star trails add up in the final picture.

If you google "olympus live composite" you can see some neat tricks done with this feature. They call it live composite because they do it in camera and with the display updating in real time... but doing it separately (and with any camera) wouldn't be bad either ^_^'

Feature proposal - Pedestrian removing photo stacking.

I have read somewhere of a photoshop function that lest you stack multiple images, and the software stacks them removing any "moving" subject which appears in different positions across a picture.

It could be used for anything like removing the pesky pedestrians in architectural shots, to airplanes from the sky... all the little annoyances that sometimes are just outside of our control and are to evident to just clone out.

I sometimes do this manually in gimp, I would go as far as saying it is fun, but it is also time consuming.
Having this in the stacking options of Photo Mate would be really neat.
Wait while more posts are being loaded