28
\$\begingroup\$

When I first saw this photo of three albatross, it seemed wrong to me somehow - like the birds were hand-drawn and added to a photo of a background. It doesn't look like an actual photograph, but I can not say what it is exactly that is giving me that feeling, except perhaps mismatches in contrast and focus in different parts of the photo.

It is from Science Magazine's December 5, 2015 This Week in Science's Not enough protection for migrating birds and it is a file photo (see the caption & credit under the photo.

I'm not asking if it is or isn't fake or modified; I'm trying to simply understand what makes it so... not-quite photographic-looking.

It reminds me of the 1950's SciFi movies where one thing is superimposed on another, pre-green screen.

enter image description here
(click for full size)

Vulnerable short-tailed albatross (Phoebastria albatrus) nest in Japan and are responding to protection measures in Japan, Canada and the USA. PHOTO: ©TUI DE ROY/MINDEN PICTURES/CORBIS

This photo can also be found at https://animalsandearth.com/featured/short-tailed-albatross-trio-at-nest-tui-de-roy.html

\$\endgroup\$
6
  • 1
    \$\begingroup\$ I may need some help with tagging; I added the [photoshop] tag as a proprietary eponym because I can't find a suitable version of [modification], nor could I find anything like a [photorealism] or [mixed-media] tag. \$\endgroup\$
    – uhoh
    Jul 26, 2023 at 9:20
  • 1
    \$\begingroup\$ Just looks like lucky with the lighting \$\endgroup\$ Jul 26, 2023 at 9:26
  • 1
    \$\begingroup\$ @blobbymcblobby lighting strikes twice (see what I did there?) \$\endgroup\$
    – uhoh
    Jul 26, 2023 at 9:27
  • 2
    \$\begingroup\$ It actually looks like a combination of things, flash fill light, over sharpening of edges via software, lasso and erase and also some hand drawing. It's fairly common practice unfortunately! Most likely, the photographer just processed and submitted. There are other pics similar to this. I am guilty of all this at times too. Bet he never expected for his image to be scrutinised by a bunch of photography geeks! 😂 \$\endgroup\$ Jul 27, 2023 at 15:26
  • 2
    \$\begingroup\$ Search for other albatross images. It seems they have very smooth feathers which looks like lassoed. \$\endgroup\$ Jul 29, 2023 at 12:42

10 Answers 10

24
\$\begingroup\$

IMO, the primary cause of the effect is differential lighting. The birds and grass have hard bright frontal light from high right. Whereas the foreground and background appear to have softer/dimmer lighting with less directionality. This can happen naturally; but when the lighting doesn't match it can seem like the images were taken in different places or at different times. Many people will notice this subliminally; and it is one of the major clues that an image has been edited without enough attention to the details.

And the fact that there is not much detail in the birds, where you would kind of expect there to be detail, gives the impression that they could be fake. Especially since there appears to be more detail in the grass and foreground. This is also a characteristic of flat/frontal lighting and monotone colors (e.g. white feathers)... the light fills in all shadows; and without shadows there is no texture/detail visible. The smaller the details (shadows) are, the more likely they are to be filled in.

\$\endgroup\$
1
  • \$\begingroup\$ Regarding the second paragraph: I think it is the lack of detail in the birds combined with their over-sharpened and very contrast contour. \$\endgroup\$ Jul 28, 2023 at 12:50
21
\$\begingroup\$

Late addition
To me, the jury's still out on whether it's fake, but I do come down on the side of 'real, but worked up [too] hard'.
In the photographer's defense, I do get the feeling the guy only got one shot [or batch of shots off an auto-trigger] & the camera got the light wrong. It's borderline over-exposed, losing detail from the lighter areas. I've done that myself often with white bird, dark background, your backdrop looks great, bird is sharp… but little feather detail left. So here, heavy cloud, sun peeks through, auto-exposure goes to hell.

In post from a RAW [idk about other cameras but] Nikon has a 'highlight protect' option that attempts to pull back some detail where it's nearly gone. It's pretty good at it, better than anything else I've tried, but it's not a miracle worker.

Maybe it was the only frame where they were posed, and not back of head, yawning, scratching under wing, crapping on grass, squabbling… that mars many a wildlife shot.

I also think it's slightly back-focussed. It's hard to tell after so much post on it, but I think the birds' breasts are closer than 'sharp', making the white feather issue even harder to try fix.


It looks like it may have had a bit of HDR done to it, to balance up the lighting. This often produces a bit of a halo around sharp edges where there was a lot of change in the light levels.

enter image description here

It can give a bit of an 'uncanny valley' look to things - in this case it almost looks like studio lighting, which in other circumstances I'd put down to something like Apple's 'portrait mode' which can do that in-camera.

See also: Halo in pictures taken with HDR

Checking a higher res preview from the linked origin, this looks even more like aggressive over-processing more common to a phone. It looks like as well as being HDR'd, it's been aggressively sharpened & de-noised, which gives a very 'plastic' look to things.

enter image description here

An example of this over-sharpen, over-de-noise effect.
First the original - small crop of a photo taken from too far away for the lens/camera to really cope, so it's noisy & not properly sharp.

enter image description here

After massive over-processing

enter image description here

\$\endgroup\$
3
  • 5
    \$\begingroup\$ It's exactly this for me. It looks like the birds are photoshopped into the picture because the edges are unnaturally sharp. \$\endgroup\$
    – DonQuiKong
    Jul 26, 2023 at 18:56
  • 4
    \$\begingroup\$ Aggressive overprocessing seems to be a common complaint with iPhones ever since the iPhone 12 came out. \$\endgroup\$ Jul 26, 2023 at 21:08
  • \$\begingroup\$ I don't think it's fake, the feet are too well integrated into the grass \$\endgroup\$
    – DonQuiKong
    Jul 28, 2023 at 21:24
16
\$\begingroup\$

Complementing Steven Kersting´s answer:

The photo looks like a diorama.

Look at the images at this google search

enter image description here

You put some stuffed animals with a printed photograph as a background, then you add some spotlights.

This happens when you add a flash to a photo taken with a natural light background. It is a really cool effect for portraits.

I feel that the photographer used a flash on a really cloudy day. I think the shadows are not parallel.

enter image description here

It can also happen when there are stormy clouds in the background and some clear sky in the front. It is not the case in this image because you can see a few clouds; they are oddly dark.

Assuming both the birds and the clouds are white, and illuminated by the same Sun (not on Tatooine) they should look similar in brightness.

\$\endgroup\$
2
  • \$\begingroup\$ Yes exactly! That is just what it's reminding me of; all those natural history museums I was taken to as a kid a million years ago... \$\endgroup\$
    – uhoh
    Jul 26, 2023 at 20:19
  • 3
    \$\begingroup\$ Interesting that most of the other answers focused on pixel peep. \$\endgroup\$
    – Rafael
    Jul 26, 2023 at 21:36
7
\$\begingroup\$

The very sharp edges of the birds that totally lack loose feathers are indeed surprising. But if you go to the site and check the close-ups, you see some lose feathers on the back of the head of the left bird. In fact the answer is on the head of the middle bird, that shows a halo:

enter image description here

... which means that at least some parts of the image have been overly sharpened, while the picture isn't that sharp to begin with (check the plumages).

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Original pic probably had a white sky, I guess the editor lassoed the sky and made it blue (or used some sky fixing software) \$\endgroup\$
    – tallberg
    Jul 27, 2023 at 6:05
5
\$\begingroup\$

I am not sure if you are seeing the same as I, but I agree, that the image at at least on a first glance looks odd.

What I believe confuses my brain is that the birds all have a much brighter front (especially front of neck and breast) than the rest of their plumage. That gives the impression that all birds are lit from the front, although they face in different directions and that the light seemingly used to illuminate the birds does not match the direction of the shadows the birds are throwing on the ground.

\$\endgroup\$
5
\$\begingroup\$

The birds have "built-in" shading.

I think this is a very important point, and I haven't seen it mentioned yet.

The gradual transition between dark and light feathers can look like it is caused by lighting and not by the natural, ingrained colors of the birds. The shading makes it look like the birds are grouped around a very bright, omnidirectional light source. Of course this light source doesn't exist, which I think makes it hard for our brains to make sense of.

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Albatrosses in general have a lot of really interesting variations on this, half of them look like they put a lot of effort into their eyeshadow every morning (eg markavery.info/2019/07/06/tim-melling-black-browed-albatross) \$\endgroup\$
    – llama
    Jul 27, 2023 at 19:27
  • \$\begingroup\$ @llama beautiful! Now I want to see an albatross, so I googled "Taiwan albatross" and found they aren't very accessible to the public :-) \$\endgroup\$
    – uhoh
    Jul 28, 2023 at 9:45
  • 1
    \$\begingroup\$ @MaxD yes now that you mention it, it really does look like that to me as well. \$\endgroup\$
    – uhoh
    Jul 28, 2023 at 9:47
4
\$\begingroup\$

I believe there are two primary causes of the uncanny valley feel:

  • The very smooth silhouette of the three birds, which makes it look like their feathers are only textures on smooth surfaces.
  • The fairly uniform and bright reflected light on the underside of the birds looks like the "ambient lighting" effect used in real time graphics to inexpensively emulate indirect lighting. In particular, the undersides of the birds are brighter than the shadows cast by the same birds, which is an unusual phenomenon in real life, but less unusual in computer graphics.
\$\endgroup\$
2
\$\begingroup\$

Two main reasons

This photograph is taken with a very powerful zoom, if you look closely at both the grass directly in front of the birds and the grass directly behind them, both appear out of focus, highlighting the main figure of the birds. Any photograph of an object in an environment taken with a powerful zoom will make the object appear cut out precisely because of the blur of the surrounding elements.

enter image description here

See Nature at mega zoom at bryanpfeiffer.com

Secondly, the characteristic of the photographed object, in particular both the plumage and the shadow of the grass, form a dark outline, better defining the figure.

enter image description here enter image description here enter image description here

\$\endgroup\$
10
  • 1
    \$\begingroup\$ Link added with more info \$\endgroup\$
    – Danielillo
    Jul 26, 2023 at 9:48
  • 3
    \$\begingroup\$ A long lens will appear to 'compress' the distances, but won't provide the halo or over sharpen/denoise. \$\endgroup\$
    – Tetsujin
    Jul 26, 2023 at 9:49
  • \$\begingroup\$ That's true. JPG artifact? \$\endgroup\$
    – Danielillo
    Jul 26, 2023 at 9:55
  • 2
    \$\begingroup\$ No. The original photo still shows the halo. It's all just been massively over-processed. To be fair to the photographer, if you only get one shot at something like this, you're gonna try your best to make it useable ;) \$\endgroup\$
    – Tetsujin
    Jul 26, 2023 at 9:59
  • \$\begingroup\$ If you look at the photos of the link that I put in the question, you can see the halo in some, for example in the lower part of the seagull \$\endgroup\$
    – Danielillo
    Jul 26, 2023 at 10:25
2
\$\begingroup\$

To substantiate earlier answers on perception of the true color image, one complement the analysis with image statistics as a tool less prone to personal bias.

This attempts a synthesis of the contribution by the answers by @Tetsujin (because of the provision of a prior/after processing illustration) in comparison to the one of the squirrel (answer by @Danielillo) and the preview of your question.

Art aside, photography equally is a measurement; and every measurement has some random error. Inconsistencies in processing (electronic) photos can yield a non-uniform error level across a JPEG image which as such can be visualized like a map. One of the programs accessible to the public I'm aware of is forensically by Jonas Wagner created around 2012 (blog post, and 12 min youtube tutorial; both in English).

Let's submit the two versions of the bird image by @Tetsujin to perform an Error Level Analysis. To highlight the differences, an alternating gif of the two annotated illustrations:

enter image description here

(analysis based on the .png files shared, vide infra)

The differences might be subtle: watch the areas of patches of similar (false) color which tend to be more uniform for the smoothed illustration. For training purpose the same was applied on the portrait of the squirrel. We observe two level of noise here: the background, on one hand vs. animal and tree on the other. This is easy to comprehend by the use of a tele lens; here, used to «get close enough» (and in classical portrait photography, to separate the object from the less interesting background which should not distract too much from the former).

enter image description here

(based on the .png shared in the answer)

enter image description here

(based on the .jpg from the author's web page here)

The comparison shows that the file format can influence the result of this analysis (after all, .png are more suitable for screen photos anyway).

And eventually, the albatross. Bodies and necks, but not (so much) the feet clearly stand out this much in comparison to the surrounding grass that a high level of editing is likely. And the sky in the background.

enter image description here

(based on the .jpg of the preview here).

The similar tool fakeimagedetector offers a slider to show the true color image/false color representation of the noise map. After some time, you can't unsee the variation in coarseness in the later.

enter image description here

(screenphoto from https://www.fakeimagedetector.com/)

Addition

There are multiple types of errors which add up to noise in the image.

  • photon shot noise: light is both wave and particle, but the arrival of photons on the detector is noncontinuous and random. The emission of photons need not be homogeneous across a surface area inspected, either.
  • electronic noise of the readout: recording an image is assigning a pixel an intensity value. To obtain a larger number of charges per pixel eventually easier to process, this amplification may introduce an error.(1)
  • quantization noise of the readout: the (amplified) analog signal is converted into a digital one. (Your analog signal may end up in the wrong digital bin.)

Aside from these, the pixels within a camera chip are not uniform; right from production, some may be less sensitive, than others. Repeated overexposure/too high levels of gain (especially ICCD camera), etc. may lead to dead pixels no longer recording a signal. Second, a pixel in operation can report intensity even in absence of a radiation source adding dark current noise. This however is more relevant to intentional observation of dim scenes / experiments with an extended period of exposure (e.g., fluorescence microscopy, astrophotography, X-ray crystallography) where one intentionally awaits temperature stabilization of the area detector, and collects dark frames for a subsequent background collection on individual frames. See (3) the above paraphrases.

These influences are all ahead of storing the information as .csv, .pgm, plain/LZW compressed .tiff. Or before the camera equally corrects white balance, adjusts sharpness, starts other electronic/AI filters (modern cellular phones) and applies a lossy compression like JPEG for the file available for you.(4) Subsequent edits on JPEGs require to encode parts of the retouched areas which then distort the file further, this and the transitions is spot by the program.

footnotes:

(1) With ICCD cameras,(2) this gain actually can be adjusted and can introduce additional problems down the road (leaving the linear range / where recorded signal is proportional to the intensity).

(2) Without particular endorsement, e.g. LaVision.

(3) Youtube videos Microscopy: Cameras and Detectors I: How Do They Work? and Microscopy: Cameras and Detectors II: Specifications and Performance by Nico Stuurman as part of a whole class around light microscopy and relevant image processing.

(4) How are Images Compressed? by BranchEducation is an instructive video about the steps of JPEG compression (color space conversion, chrominance downsampling, discrete coside transform, quantization the eventual Huffman encoding). It both explains the steps, and one gets an idea why recording a photo in a raw format rather than JPEG can be advantageous. (Not every image format constrains the the intensity depth to 8 bit per (color) channel (i.e., 256 levels) only.)

\$\endgroup\$
8
  • 1
    \$\begingroup\$ What does "error" mean here? how is error measured without some sort of ground truth image? \$\endgroup\$
    – Vaelus
    Jul 28, 2023 at 0:44
  • \$\begingroup\$ @Vaelus this is image statistics - Buttonwood can answer more accurately, but it's something like the difference between a pixel's value and perhaps the average of the four adjacent pixels. Smooth gradients will be nearly zero, hot pixels will be super bright, and the ever-present noise (shot noise and other effects) will result in the "snow" we see in these images. You might imagine this being an ultra-high frequency high-pass filter. \$\endgroup\$
    – uhoh
    Jul 28, 2023 at 1:15
  • \$\begingroup\$ @Vaelus images from different sources will often have different noise levels so if they are both present in one combined image they'll stand out in a noise image. \$\endgroup\$
    – uhoh
    Jul 28, 2023 at 1:16
  • 1
    \$\begingroup\$ @uhoh I complemented the answer with a section what typical errors in an image (statistically speaking) can be, and some references. Because of some previous exchanges with you more related to crystallography, some of them might be equally of interest there (e.g., slowly sweeping $\phi$ on a single crystal X-ray diffractometre, recording with a CCD [instead of a point detector], and even more, with an image plate), too. \$\endgroup\$
    – Buttonwood
    Jul 31, 2023 at 9:36
  • 1
    \$\begingroup\$ @Buttonwood I'm following your post and I find it really interesting, and now with the edit I'll start again and go through carefully and check all your links, it demands no less! Thanks for pursuing this so thoroughly! In another life I also looked at long exposure photographic plates of the stars :-) We never got out the dry ice to reduce reciprocity failure, but we had some fun! \$\endgroup\$
    – uhoh
    Jul 31, 2023 at 11:11
1
\$\begingroup\$

Note really an answer perhaps, more of an opinion:

For me it seems like the theres a lot of fuss over not very much.

Yes, there is much pixel peeping at edges, level of sharpening used, but I do not think that that was defining it as 'looking fake'. Sure, the processing looks like a lot on close inspection but I do not think that that 'is it' in terms of whats going on, if anything is...

Its simply the brightness of the birds against the duller background that is making people think it is not an entirely natural image. But I've seen, or even taken images that have given that effect before, usually unintentionally, and usually in natural lighting conditions where the sun pokes through on a not entirely sunny day and the subjects are bright anyway.

I think someone else did touch on it - the plumage of the birds are somewhat naturally luminous, and catch available light in a very broad way:

enter image description here

I do think the shadows and lighting match the birds self shadow and the fall on the ground

enter image description here

There is bright sunlight, it is falling on everything in the frame, the background and the birds.

Perhaps there is expectation that the dark plumage would also go bright with the fall of sunlight on it and I think the plumage does not do that, it retains its darkness.

Due to the natural brightness picked up on the birds, it is this area that looks a little too bright for people, and that is where the suggestion of an artificial light comes in, but i would argue against that, not last because using an artificial light is not a great idea with bird wildlife photography - i thin kit would be frowned upon in such circles, and the photographer has done a fair bit of bird photography. Flash is even less likely, as is a reflector - none of those ideas really match what is going on here, which is that the photographer was lucky in picking right time, place, and lens for the frame.

Coming from another angle, i have a background in 3d cgi as well, and to get that look in basic surfacing, we would modify fall off, making it broad, and probably adjusting certain bounce/diffuse maps and colors of the birds - all of which would produce a brighter, broader lit, bird, without adding an additional light - the sun is enough.

What makes it look unnatural?

A combination of lighting conditions at the time, direction, place, and lens, and the birds plumage,, against the background, is what makes them look superimposed. Yes, there is probably a bit of an issue with aggressive post processing, but I do not believe that this is the singular issue that is creating an issue.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.