cameras.....are not what you thought they do.

[ocnbrze]

its crazy on how much software plays a role on how a picture will look on phones these days. if you have the time check out these videos by Marques Brownlee

this video says that the photo of the moon taken on your s23 ultra is fake.......crazy

here is the reddit thread talked about in MB's video:

https://www.reddit.com/r/Android/comments/11nzrb0

and yet another article thats been translated:

달 촬영

이서영 [ 목차 ] 1. 달 촬영 개요 2. 달 인식 엔진을 위한 데이터 학습 과정 3. 달 밝기 제어 과정 4. 달 촬영 손떨림 제어 과정 5. 달 디테일 개선 엔진의 학습 과정 6. 달 촬영 과정 [ 달 촬영 개요 ] 갤럭시는 S10부터 카메라에 AI 기술을 적용하여, 사용자가 시간과 장소에 구애받지 않고 최고의 사진을 촬영할 수 있도록 노력하고 있습니다. 이를 위해, AI가 찍고자 하는 대상을 인식하여 최적의 결과물을 도출할 수 있도록 도와주는 Scene...

r1-community-samsung-com.translate.goog

its just crazy stuff.....i am not sure how i feel about it. i guess for me as long as the pictures i take look good who is to tell me otherwise, right?

 

[olbriar]

Could be but I'm not completely buying into Samsung doctoring pics. Below I post the pic I took of the moon with my s21 ultra 100x. Note the tree tops that I could not avoid and captured in the image.

 

[nickdalzell]

I'm plenty satisfied so long as the camera takes a good enough photo. I'm even happy with the results from a Note II or S4 Mini. However, lately the only uses I have for a smartphone camera are for landscapes (desktop wallpapers) or reference shots at work prior to disassembly. I wish I had something or someone far more important to take a photo of where perhaps, a modern phone camera would be best, but alas, they are all dead.

I have always been of the view "it's not the camera, it's the photographer". You can, with enough skill, get great photos with a Brownie camera if you like, and it's a great skill to acquire. Equally, you can, with the right tweaks, get as good a photo from a HTC Evo as you would from an S20 Ultra.

 

[ocnbrze]

Could be but I'm not completely buying into Samsung doctoring pics. Below I post the pic I took of the moon with my s21 ultra 100x. Note the tree tops that I could not avoid and captured in the image.

so the ai in samsung's camerq actually will detect anything in front of the moon and impose the moon behind the object. another youtuber also tried his own experiment and was able to add a picture into the moon. basically he dialed down the image so it's blurred and then added a picture of his face. the sammsung camera then added the picture into the moon somehow.

so far it is only the moon that it is done as samsung has used this to advertise their super zoom features as described in the above video.

apple is doing the same thing and supposedly Huawei has done it as well. its crazy how ai is now invading our lives and we don't even know it.

 

[Milo Williamson]

A little off topic but today I manage to snapshot a Dog's tail, (Ring around the sun) with just my moto and I will upload it tomorrow

 

[nickdalzell]

We know AI is invading our lives. Pretty much anyone has an Alexa or Google speaker, or uses the Assistant. The problem is that nobody sees it as a threat, just that it's the future

Yeah, right into a future dystopia that many movies and books warned us about! More recently Microsoft tried releasing an AI chatbot called Tay, but that quickly devolved into her being racist, anti-human and praising Hitler. Heck, they forecasted the dangers of AI as far back as the 1960s, with 2001: A Space Odyssey

 

[Milo Williamson]

 

[jack58]

Could be but I'm not completely buying into Samsung doctoring pics. Below I post the pic I took of the moon with my s21 ultra 100x. Note the tree tops that I could not avoid and captured in the image.
View attachment 165975

I agree with you. Also the trees add to your photo in my opinion.
Like this one I took a couple years ago with my Nikon

 

[Davdi]

Here's one I took a while ago with my Fuji bridge camera.

 

[Milo Williamson]

ME! About a few days ago, mine seem working pretty well.

 

[Hadron]

We know AI is invading our lives. Pretty much anyone has an Alexa or Google speaker, or uses the Assistant. The problem is that nobody sees it as a threat, just that it's the future

Yeah, right into a future dystopia that many movies and books warned us about! More recently Microsoft tried releasing an AI chatbot called Tay, but that quickly devolved into her being racist, anti-human and praising Hitler. Heck, they forecasted the dangers of AI as far back as the 1960s, with 2001: A Space Odyssey

"AI" like HAL is not a threat currently - we're frankly no nearer that type of genuine general intelligence than we were when 2001 was released. The bigger threat is the ability to fake images, videos, voices and text accelerating the trends that we've already seen from social media and information (or disinformation) bubbles. Even without the deliberate use of these things to manipulate and deceive there have already been many examples of people using these chatbots to research topics and being presented with fake articles and fake sources. How many people of a conspiracist bent - something that social media algorithms already feed - will conclude when the source of an argument that appealed to them cannot be found not that they were using a fatally flawed tool (ChatGPT, Bard or the rest) that made the whole thing up, but that "they" have censored it to stop people learning the truth? And conversely awareness of the ease of producing fake sources and evidence risks the same outcome, of people rejecting real evidence as fake because they know there are so many fakes (and thus tending to accept things that fit their own biasses)?

No, I'm not worried about HAL or Skynet. More prosaically, I'm worried about the capacity of these much, much more limited tools we have to accelerate the worst trends that have already developed. In fact if these things had an understanding of reality, of the concept of reality, and some capacity to form judgement, I might actually be less worried (except that the corporations behind them would invariably build in biases to favour their interests...).

 

[mikedt]

We know AI is invading our lives. Pretty much anyone has an Alexa or Google speaker, or uses the Assistant. The problem is that nobody sees it as a threat, just that it's the future

Yeah, right into a future dystopia that many movies and books warned us about! More recently Microsoft tried releasing an AI chatbot called Tay, but that quickly devolved into her being racist, anti-human and praising Hitler. Heck, they forecasted the dangers of AI as far back as the 1960s, with 2001: A Space Odyssey

I'm sure the problem with these bots, like Alexa, Siri, Bixby, Tay, Chatgpt, Bard, or whatever, is they have no actual concept of true reality and history. If Tay was told that Hitler was a nice man, then it likely will praise him.

 

[Milo Williamson]

I just read an intresting article of "We are even closer by of having an anime charater, now these days, who wants our SS number and cell phone."