Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Can you tell if an image is AI-generated?

aiquiz.ronsor.com

Post reply on HN
Print view View original

101–107 of 107 posts

Re: Show HN: Can you tell if an image is AI-generated?

#101

Post by gus_massa »

ronsor wrote: post #97
cal85 wrote: post #78

To go through all that and not be shown the answers at the end is beyond infuriating.

If you save your email, full results will be emailed next week.

I tried this morning (19 of 30 (63%)). I just tired again, and clicked "Continue Quiz!" and then "Save for later... [arrow]" and it offered the possibility to add my email. So the GP can still fill an email and get the result next week. (With the real email or mailinator.)

Perhaps you should change the text form "Save for later... [arrow]" to something like "Get the results by email... [arrow]".

Re: Show HN: Can you tell if an image is AI-generated?

#102

Post by dec0dedab0de »

Xelynega wrote: post #96
dec0dedab0de wrote: post #73

Earlier quoted context omitted.

Filtered is the process of removing parts from the whole. Like how a water filter removes unwanted chemicals. Selected means it was chosen. So after filtering you could make a selection from all or part of the remaining choices. Cherry picked, means selected individually to meet a goal. Selecting very bad results to be removed is almost the opposite of cherry picking. Cherry picking would be picking only the best ima…

> picking only the best images and presenting them > they removed the worst images and showed the rest Is removing rejects not a way to pick only the best?

Is removing rejects not a way to pick only the best?

It is not, unless your rejection criteria is "everything that is not the best." However, if your rejection criteria is "not meeting basic requirements" then you are left with the best, and the second best and the middle, all the way down to the bottom of adequate.

Re: Show HN: Can you tell if an image is AI-generated?

#106

Post by kazinator »

Very tough calls on art, or scenes without a lot of shadow and light and geometric objects, like certain nature scenes. I tended to err toward AI.

  You correctly classified 22 of 30 (73%) images!

  You identified 13 of 14 (92%) AI-generated images.

  You identified 9 of 16 (56%) human-made images.
I would be more reliable between "photo versus AI", for certain kinds of photos where you can look for certain features being wrong.

Re: Show HN: Can you tell if an image is AI-generated?

#107

Post by Xelynega »

dec0dedab0de wrote: post #102
Xelynega wrote: post #96

Earlier quoted context omitted.

> picking only the best images and presenting them > they removed the worst images and showed the rest Is removing rejects not a way to pick only the best?

Is removing rejects not a way to pick only the best? It is not, unless your rejection criteria is "everything that is not the best." However, if your rejection criteria is "not meeting basic requirements" then you are left with the best, and the second best and the middle, all the way down to the bottom of adequate.

So if "basic requirements" in this case are "on first glance like a face good enough to fool a human", I don't see how you could have a "best" criteria that's significantly different without accounting for the odd glitches in the current set.

You're making a distinction that I understand can exits. I'm just failing to see where it exists in this case.

Post reply on HN