Samsung “space zoom” moon shots are fake, and here is the proof

Imagine this future:<p>Sensor quality in phones goes down, AI makes up for it because good sensors are expensive, but compute time in the cloud on Samsung owned servers is cheap. You take a picture on a crappy camera, and Samsung uses AI to &quot;fix&quot; everything. It knows what stop signs, roadways, busses, cars, stop lights, and more <i>should</i> look like, and so it just uses AI to replace all the textures.<p>Samsung sells what&#x27;s on the image to advertisers and more with the hallucinated data. People can&#x27;t tell the difference and don&#x27;t know. They &quot;just want a good looking picture&quot;. People further use AI to alter images for virtual likes on Tiktok and Insta.<p>This faked data, submitted by users as &quot;real pics in real places&quot; is further used to train AI models that all seem to think objects further away have greater detail, clarity, and cleanliness than they should.<p>You look at a picture of a park you took, years before, and could have sworn the flowers were more pink, and not as red. You are assured, by your friend who knows it all, that people&#x27;s memories are fallible; hallucinating details, colors, objects, sizes, and more. The image, your friend assures you further? &quot;Advanced tech captured its pure form perfectly&quot;.<p>And thus, everyone will demand more clarity, precision, details, and color where their eyes don&#x27;t remember seeing.

I think they&#x27;re probably right about the AI-sharpening using specific knowledge about the moon... However, they are wrong about the detail being gone in the gaussian-blurred image.<p>If they applied a perfect digital gaussian-blur, then that <i>is</i> reversible (except at the edges of the image, which are black in this case anyway). You still lose some detail due to rounding errors, but not nearly as much as you might expect.<p>A gaussian blur (and several other kinds of blur) are a convolution of the image with a specific blur function. A convolution is equivalent to simply multiplying pointwise the two functions in frequency space. As long as you know the blur function exactly, you can divide the final image by the gaussian function in frequency space and get the original image back (modulo rounding errors).<p>It is not totally inconceivable that the AI model could have learned to do this deconvolution with the Gaussian blur function, in order to recover more detail from the image.

Hey all, it&#x27;s the author of the reddit post here. First of all, let me say that I don&#x27;t usually frequent HN, but the comments on here are of such high quality, that I might need to change that. I got semi-depressed on reddit, with people misattributing statements and, in general, not being overly, uh, skeptical :)<p>That being said, there were a few comments on here about gaussian blur and deconvolution, which I would like to tackle. First, I need to mention that I do not have an maths&#x2F;engineering background. I am familiar with some concepts, as I&#x27;ve used deconvolution via FFT several years ago during my PhD, but while I am aware of the process, I don&#x27;t know all the details. I certainly didn&#x27;t know that the image that was gaussian blurred could be sharpened perfectly - I will have to look into that. In fact, I used gaussian blur to redact some private information (like in screenshots), and it&#x27;s very helpful to know if I haven&#x27;t redacted anything and the data is recoverable. Wow.<p>I would love to learn more about the types of blur that cannot be deconvoluted.<p>However, please have in mind that in my experiment:<p>1) I also downsampled the image to 170x170, which, as far as I know, is an information-destructive process<p>2) The camera doesn&#x27;t have the access to my original gaussian blurred image, but that image + whatever blur and distortion was introduced when I was taking the photo from far away, (whatever algo they are using doesn&#x27;t have access to the original blurred image to run a perfect deconvolution on)<p>3) Lastly, I also clipped the highlights in the last example, which is also destructive (non-reversible), and the AI hallucinated details there as well<p>So I am comfortable saying that it&#x27;s not deconvolution which &quot;unblurs&quot; the image and sharpens the details, but what I said - an AI model trained on moon images that uses image matching and a neural network to fill in the data.<p>Thank you again for your engagement and your thoughtful comments, I really appreciate them, and have learned a lot just by reading them!

I took a max-(non-optical)-zoom photo of a rabbit in my yard a while back using an iphone, then further enlarged the result to see how it did - in the details it looked like an impressionistic painting of a rabbit, facing the camera and looking left. The actual rabbit was looking away from the camera and to the right. The eye and face were not visible.<p><a href="https:&#x2F;&#x2F;i.ibb.co&#x2F;Kz7Sbm2&#x2F;8-EA85-C12-5-B11-44-D8-9566-461-C98627-E10.jpg" rel="nofollow">https:&#x2F;&#x2F;i.ibb.co&#x2F;Kz7Sbm2&#x2F;8-EA85-C12-5-B11-44-D8-9566-461-C98...</a>

While I understand what the author tries to say, I have to point out that ship has long sailed. Samsung just pushed it a bit too far and slapped a &quot;scene optimizer&quot; label on it.<p>AI has been used in &quot;cell phone photography&quot; for a few years, at lease since Pixel 2 where a mediocre sensor produced much better pictures than what people expect (maybe there are other players who did this even earlier). And every manufacturer started doing it, including Apple. Otherwise, do you think &quot;night mode&quot; is just pure magic? Of course not, algorithms are used everywhere.<p>How do you define &quot;fake&quot;? In podcasts, Verge editor Nilay Patel has asked various people &quot;what is a photo&quot;, because the concept of a &quot;photo&quot; has become increasingly blurry. That is the question the author is asking, and people may have different answers from the author&#x27;s.

This problem is not new. In 2019, Huawei introduced a special image processing feature in its smartphone camera app, the &quot;Moon Mode&quot; (opt-in). Missing details are added to the moon photos via machine learning inference from a pre-trained model. Huawei then started marketing these processed images as a showcase of its new smartphone&#x27;s photography performance. In China, it was widely criticized by tech reviewers [1][2] as misleading, and &quot;Moon Mode&quot; became a running gag among tech enthusiasts for a while.<p>It seems that Samsung simply adopted the same tactic to compete...<p>On the Huawei &quot;Moon Mode&quot; controversy, one can even find a research paper [3] published in a peer-reviewed social studies (!) journal, <i>Media, Culture &amp; Society</i>:<p>&gt; <i>This is where the controversy began: Chinese tech critic Wang’s (2019) posting on Weibo, the Chinese equivalent of Twitter, made quite a splash. In his post, Wang put forward a shocking argument: he said that Huawei’s Moon Mode actually photoshops moon images. He contended that, based on his self-conducted experiments, the system ‘paints in pre-existing imagery’ onto photographed takes, re-constructing details that are not captured in the original shots. Huawei immediately refuted these claims, stressing that the Moon Mode system ‘operates on the same principle as other Master AI modes that recognize and optimize details within an image to help individuals take better photo&#x27; </i><p>[1] <a href="https:&#x2F;&#x2F;www.androidauthority.com&#x2F;huawei-p30-pro-moon-mode-controversy-978486&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.androidauthority.com&#x2F;huawei-p30-pro-moon-mode-co...</a><p>[2] <a href="https:&#x2F;&#x2F;www.phonearena.com&#x2F;news&#x2F;Is-the-Moon-Mode-on-the-Huawei-P-30-Pro-faking-parts-of-photographs_id115554" rel="nofollow">https:&#x2F;&#x2F;www.phonearena.com&#x2F;news&#x2F;Is-the-Moon-Mode-on-the-Huaw...</a><p>[3] <a href="https:&#x2F;&#x2F;journals.sagepub.com&#x2F;doi&#x2F;full&#x2F;10.1177&#x2F;01634437211064964" rel="nofollow">https:&#x2F;&#x2F;journals.sagepub.com&#x2F;doi&#x2F;full&#x2F;10.1177&#x2F;01634437211064...</a>

This is just make hipsters get into old point and shoot digital cameras... good thing I kept my Canon A540.<p>All the subtle trickery manipulation that the smart phone&#x27;s doing to reality is concerning. Smoothing people&#x27;s faces, making their eyes pop, enhancing the shit out of the colours, and now plopping fake objects overtop of the real ones.<p>Future concerns of this technology should range from a low-key disconnect from reality, to the complete inability to photograph certain objects or locations.<p>Imagine dusting off a 30 year old digital camera, finding some AA batteries to put in it, snap a selfie and then realizing just how ugly we all are and how washed out the polluted world actually looks without a bunch of narcissism-pandering enhancements.

&gt; I downsized it to 170x170 pixels and applied a gaussian blur, so that all the detail is GONE. This means it&#x27;s not recoverable, the information is just not there, it&#x27;s digitally blurred<p>Strictly speaking, applying a Gaussian blur does not destroy the information. You can undo a Gaussian blur with a simple deconvolution, which is something I would expect even a non-AI image enhancement algorithm to do (given that, you know, lenses are involved here).<p>I&#x27;d like to see what detail can be &quot;recovered&quot; with just the downsizing, which DOES destroy information.

OP here again.<p>I photoshopped one moon next to another (to see if one moon would get the AI treatment, while another would not), and managed to coax the AI to do exactly that.<p>This is the image that I used, which contains 2 blurred moons: <a href="https:&#x2F;&#x2F;imgur.com&#x2F;kMv1XAx" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;kMv1XAx</a><p>I replicated my original setup, shot the monitor from across the room, and got this: <a href="https:&#x2F;&#x2F;imgur.com&#x2F;RSHAz1l" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;RSHAz1l</a><p>As you can see, one moon got the &quot;AI enhancement&quot;, while the other one shows what was actually visible to the sensor - a blurry mess<p>I think this settles it.

Famously, upscaled Obama: <a href="https:&#x2F;&#x2F;twitter.com&#x2F;Chicken3gg&#x2F;status&#x2F;1274314622447820801" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;Chicken3gg&#x2F;status&#x2F;1274314622447820801</a><p>I picture someone 20 years from now trying to find out what their parent really looked like when they were young. The obviously smoothed-out face filters are already giving way to AI-powered homogenization. And the filtering is moving deeper down the stack from the app to the camera itself. There will be no &quot;original&quot;.

This issue has been known for a few years. Here&#x27;s a more thorough analysis from January 2021:<p><i>&quot;Is the Galaxy S21 Ultra using AI to fake detailed Moon photos?&quot;</i>: <a href="https:&#x2F;&#x2F;www.inverse.com&#x2F;input&#x2F;reviews&#x2F;is-samsung-galaxy-s21-ultra-using-ai-to-fake-detailed-moon-photos-investigation-super-resolution-analysis" rel="nofollow">https:&#x2F;&#x2F;www.inverse.com&#x2F;input&#x2F;reviews&#x2F;is-samsung-galaxy-s21-...</a><p>Note that the author decides that Samsung&#x27;s photos are &quot;not fake&quot;, in the sense that they were not doctored after being taken with the phone. However, the article decisively proves that they&#x27;re being heavily doctored in-camera.<p>Another test would be to shoot RAW + JPEG if the camera supports it. A true RAW image would reveal what the sensor is actually capturing.

Pentax K3-II 300mm: <a href="https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1010562706237038633&#x2F;1082113518809133138&#x2F;IMGP8315.jpg" rel="nofollow">https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1010562706237038633&#x2F;1...</a><p>Sensor: 23.5 x 15.6mm 24MP<p>S21 Ultra: <a href="https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1010562706237038633&#x2F;1039675963270758480&#x2F;20221108_175904.jpg" rel="nofollow">https:&#x2F;&#x2F;cdn.discordapp.com&#x2F;attachments&#x2F;1010562706237038633&#x2F;1...</a><p>Telephoto sensor: 3.3 x 4.3mm 10MP (240mm equivalent)<p>Compare side by side: <a href="https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;QwnV99D" rel="nofollow">https:&#x2F;&#x2F;imgur.com&#x2F;a&#x2F;QwnV99D</a>

I&#x27;m not very convinced. It looks like they are doing some processing, maybe special to the moon but it looks more like some form of sharpening or contrast boosting than adding detail. In all of the examples it seems that there is information in the original (dark spots) that are getting boosted.<p>It would be interesting to see this tried on a source image that isn&#x27;t the moon. Just white with a few dark spots. Does it actually add in completely new craters, or just where there are existing smudges? Or do something like half of a moon photo and have white, does it add craters to the white side?<p>The OP tried to do this by changing the contrast but I failed to see any craters appearing where there wasn&#x27;t already dark spots in the source photo.<p>It does seem strange that the OP is using an image of the moon to start and that they don&#x27;t provide a still shot of the one where they modified the brightness levels to cause clipping. It doesn&#x27;t really &quot;drive the point home&quot; as claimed.<p>Of course the answer to these may be that you need something moon-like enough to trigger the moon optimizations. But if that is the answer it would be interesting to see something that comes right up to the threshold where it either snaps in and out of these optimizations or two very similar images produce widely different results.

Does it enhance the moon only, or the Taj Mahal, Grand Canyon and Montblanc, too?<p>How long until your phone detects the subject based on geo localization and replaces your shot with a stock image of the subject from an immense database of selected, professional-looking pictures?

Officially documented by Samsung in Korean:<p><a href="https:&#x2F;&#x2F;r1.community.samsung.com&#x2F;t5&#x2F;camcyclopedia&#x2F;%EB%8B%AC-%EC%B4%AC%EC%98%81&#x2F;ba-p&#x2F;19202094" rel="nofollow">https:&#x2F;&#x2F;r1.community.samsung.com&#x2F;t5&#x2F;camcyclopedia&#x2F;%EB%8B%AC-...</a><p>Ongoing arguments over what this counts as:<p><a href="https:&#x2F;&#x2F;old.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11or39c&#x2F;samsungs_algorithm_for_moon_shots_officially&#x2F;" rel="nofollow">https:&#x2F;&#x2F;old.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11or39c&#x2F;samsungs_a...</a><p><a href="https:&#x2F;&#x2F;twitter.com&#x2F;search?q=samsung%20moon" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;search?q=samsung%20moon</a>

I guess there is some AI algorithm that does zoom as postprocessing. That AI knows the moon so it can fill in the blanks and compensate for a (relatively) crappy sensor.<p>So in the future, there are either cameras that can see what others have seen before, and those that can truly capture new, true, detail (true as in, without filling it with estimations)

Samsung is learning the best from Huawei <a href="https:&#x2F;&#x2F;www.androidauthority.com&#x2F;huawei-p30-pro-moon-mode-controversy-978486&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.androidauthority.com&#x2F;huawei-p30-pro-moon-mode-co...</a>

Idea for further testing to prove the hypothesis beyond reasonable doubt:<p>What happens if you manually add a few additional artificial details to your copy of the image -- like a trap street in maps. Does the camera slip up and show flawless moon instead?

I have a few questions: Is this specific for the moon or is the AI generally sharpening around dark areas? If it is specific to the moon then how many common objects does the camera recognize? If you did the same test with a picture of a stop sign or a popular car would the results be similar? Also is this processing done on the camera or does this require an internet connection and work on Samsungs end? I have known about AI image enhancement for years but the idea of recognizing then re-texturing common objects is something I had never considered.

Can it be just deconvolution? <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Deconvolution" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Deconvolution</a><p>The Gaussian blur they applied is theoretically reversible in a continuous function and infinite precision. In a discrete function (like a image in the computer) and only a few dozens of bits it&#x27;s not 100% reversible, but it can be partially undone and get a sharper image (that is not as sharp as the initial image).

I guess the old &quot;seeing is believing&quot; can be thrown out the window nowadays.

Oh but the plot thickens. Was this post itself a result of AI plagiarizing&#x2F;paraphrasing of another older post?<p><a href="https:&#x2F;&#x2F;old.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11nzrb0&#x2F;samsung_space_zoom_moon_shots_are_fake_and_here&#x2F;jbuet1x&#x2F;" rel="nofollow">https:&#x2F;&#x2F;old.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11nzrb0&#x2F;samsung_sp...</a>

It&#x27;s creepy how deceptive phone cameras are. I&#x27;ve noticed in pictures that I&#x27;ve taken that certain elements have been postprocessed to look better than they actually are.

Samsung is such a tasteless company.<p>This is a cool and useful tech, but it obviously needs a good marketing story to avoid looking creepy. To think that no one will notice just makes it worse.

A gaussian blur is not a good test. It does not technically remove all the information from the image. As such this test can&#x27;t distinguish between unblur or actual moon-pasting.<p>Binning the image, or cropping it in Fourier space, would be a better test.

I think it’s even more damning seeing this happen on the phone near the bottom of the article. There’s no way a phone camera sensor can capture all those raw values of the moon with 45X digital zoom with a phone telephoto lens (you can only see so far with 5 mm of packaging). From a digital sensor point of view, it probably sees a white blob and fires up the AI when it has a contrast and lighting similar to the moon

Wait ... a Gaussian blur? That doesn&#x27;t remove ANY information from the image. It looks blurred to our eyes but the same information is all there.<p>The information is absolutely not gone.<p>Does it state that it hallucinates craters that were not there in the original, or is it possible the filters simply did an FFT, adjusted the power spectrum to what we expect of a non-blurry picture, hence inverting the Gaussian blur?<p>EDIT: Note that a deblur of a smooth but &quot;noisy&quot; image can cause &quot;simulation&quot; or &quot;hallucination&quot; entirely without AI. Could be any number of things causing an output image like that (wavelet sharpening, power spectrum calibration, ...). Even if the information isn&#x27;t recoverable as such, a photo of a Gaussian blur has an unnatural power spectrum that could easily &quot;trick&quot; conventional non-AI algorithms into doing such things.<p>Especially since the only thing I see in the output is &quot;more detail&quot; (i.e. simply a different power spectrum than the author expected..)

How it is possible that people belived that these images are not artificially enchanced? Like it is obvious, there is no way a phone sensor can take such detailed photos of the Moon. I am surprised this was even a discussion.

The most impressive thing is how dumb the Reddit comments are. Incorrect factual claim after incorrect factual claim. Confidently stated and widely upvoted.<p>Was it always like this? Were <i>we, people</i> always like this?

I am not convinced. Looks like a bog standard sharpening alghorithm to me.<p>I’ve taken pictures of the moon at 100x optical zoom, and if Samsung is really faking this they’re doing a truly awful job of it.

MP3 was perceptually-aware lossy compression. How many people complain about it 20 years later

I tinkered with the idea to let a app use your personal gallery as training data and create a &quot;barney camera&quot;(himym) where 100% of photos of you and your friends would always be perfect, it would be impossible to capture a bad photo.. same here, its the moon just perfect .. who cares its not like it changes ..

Blatant but how is this different than all of the other &quot;AI&quot; touchups to things like faces?

What happens if you film a crime in the apartment far away and the AI fills in the details?

Samsung is a corrupt company with a corrupt culture.<p>Starting from the top - bribery, embezzlement, illegal transactions, stock manipulation, perjury - all the way down to their &quot;partners&quot;.<p>They&#x27;ve been caught multiple times lying in their products description.<p>I won&#x27;t add links as there are too many, but a quick search for Samsung and the keywords I mentioned will bring many results.<p>(let the downvoting begin)

So apparently iPhone and Pixel phones are not adding non existent details to Moon unlike Samsung<p><a href="https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11onztx&#x2F;uibreakphotos_sent_a_very_interesting_test&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.reddit.com&#x2F;r&#x2F;Android&#x2F;comments&#x2F;11onztx&#x2F;uibreakpho...</a>

Does DxOMark penalize such falsifications?

Every few months there&#x27;s a new clown on reddit that doesn&#x27;t know anything about photography or gaussian blur and tries to expose smartphones for using AI based sharpening and interpolation.<p>Oh nooooo this phones photos come out much sharper than my peanut brain expected!!! Fraud!!!!!!!!!!

Has anyone found a decent use of the 100x zoom on these phones? Could it be that the sheer hype it causes compensates for the extra cost of putting it on the phone? Like, even if nobody uses it it stills turn an overall profit because of the marketing

In this regard, the AI is acting just like the human brain - adding details that may or not be present in what the viewer is seeing. Details that the viewer expects to see, so the brain delivers against this expectation.

<i>Samsung Galaxy Phone - now hallucinating a more beautiful reality</i>™

Maybe I can get a selfie with my S23 and get a picture of Leo DiCaprio!

Huawei phones did this a few years back.<p>You can get sued if you publish someones real pictures without makeup and photo-shopping. AI beatified pictures have the same criteria.

we need &quot;raw&quot; camera apps for our smartfones. like oraganic food, &quot;no AI was used in the making of this photo&quot;

Do we really need cameras at this point? Remove cameras, and just ask user what he is taking photo of, then generate the image.

Am I supposed to be convinced by this? Cause I&#x27;m not.<p>It looks more like moon mode is assembling fake detail out of shot noise from a photo being taken in the dark indoors. That&#x27;s enough to cancel out his blurring and all that.<p>Even if it is guided upsampling with a &quot;this is a moon prior&quot;, so what? That&#x27;s not &quot;fake&quot;, it&#x27;s constrained by the picture you took.

Wasn’t this obvious from Day 1

It seems somebody from VW started working at Samsung

That&#x27;s not a moon, it&#x27;s a space station.

I&#x27;m surprised that some people are surprised.

tl;dr: AI is used to add detail that doesn&#x27;t exist to your photos.

So this person has been able to replicate a process that he thinks Samsung must&#x27;ve used, and this is proof? What a load of rubbish this article is!<p>Not denying that Samsung or any other brand would fake things, but this is in no way any proof at all.