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

4o Image Generation

openai.com

Post reply on HN
Print view View original

621–629 of 629 posts

Re: 4o Image Generation

#621

Post by ASalazarMX »

gloosx wrote: post #389
ASalazarMX wrote: post #352

Earlier quoted context omitted.

I think we're really fscked, because even AI image detectors think the images are genuine. They look great in Photoshop forensics too. I hope the arms race between generators and detectors doesn't stop here.

We're not. This PNG image of a wine glass has JPEG compression artefacts which are leaking from JPEG training data. You can zoom into the image and you will see 8x8 boundaries of the blocks used in JPEG compression, which just cannot be in a PNG. This is a common method to detect AI-generated image and it is working so far, no need for complex photoshop forensics or AI-detectors, just zoom-in and check for compressio…

If JPEG compression is the only evident flaw, this kind of reinforces my point, as most of these images will end up shared as processed JPEG/WebP on social media.

Re: 4o Image Generation

#622

Post by sayamqazi »

vlovich123 wrote: post #354
sayamqazi wrote: post #307

Earlier quoted context omitted.

I think there is a critical aspect of human visual learning which machine leanring cant replicate because it is prohibitively expensive. When we look at things as children we are not just looking at a single snapshot. When you stare at an object for a few seconds you have practically injested hundreds of slightly variated images of that object. This gets even more interesting when you take into account real world is…

Then explain blind children? Or blind & deaf children? There's obviously some role senses play in development but there's clearly capabilities at play here that are drastically more efficient and powerful than what we have with modern transformers. While humans learn through example, they clearly need a lot fewer examples to generalize off of and reason against.

> Then explain blind children I was only talking about vision tasks as an example. You can extend the idea to any sense.

> While humans learn through example, they clearly need a lot fewer examples to generalize off of and reason against.

Human brain has been developing over millenia. machines start from zero. What if this few example learning is just an emergent capbaility of any "leanring function" given enough compute and training.

Re: 4o Image Generation

#623

Post by Too »

vunderba wrote: post #370

Ran through some of my relatively complex prompts combined with using pure text prompts as the de-facto means of making adjustments to the images (in contrast to using something like img2img / inpainting / etc.) https://mordenstar.com/blog/chatgpt-4o-images It's definitely impressive though once again fell flat on the ability to render a 9-pointed star.

Funny how it changed the spelling in the Magic coloring book. Westher, wntility, substittutesfor, yyears. What's worse is that it removed the CO2 tank and changed some vital numbers, eg 8' to 3', completely altering the meaning of the schematic. Not what i would call passing with flying colors. Still pretty cool as a party trick, like most other AI outputs, useful only with careful review.

Re: 4o Image Generation

#625

Post by gloosx »

ASalazarMX wrote: post #621
gloosx wrote: post #389

Earlier quoted context omitted.

We're not. This PNG image of a wine glass has JPEG compression artefacts which are leaking from JPEG training data. You can zoom into the image and you will see 8x8 boundaries of the blocks used in JPEG compression, which just cannot be in a PNG. This is a common method to detect AI-generated image and it is working so far, no need for complex photoshop forensics or AI-detectors, just zoom-in and check for compressio…

If JPEG compression is the only evident flaw, this kind of reinforces my point, as most of these images will end up shared as processed JPEG/WebP on social media.

You didn't get it. The image contains ALL compression artifacts from different algorithms mashed up in a single picture, the JPEG is just prevalent.

Re: 4o Image Generation

#626

Post by ASalazarMX »

gloosx wrote: post #625
ASalazarMX wrote: post #621

Earlier quoted context omitted.

If JPEG compression is the only evident flaw, this kind of reinforces my point, as most of these images will end up shared as processed JPEG/WebP on social media.

You didn't get it. The image contains ALL compression artifacts from different algorithms mashed up in a single picture, the JPEG is just prevalent.

Oh, I see. There's still room for reliable detection then.

Re: 4o Image Generation

#627

Post by bobxmax »

hansmayer wrote: post #612
bobxmax wrote: post #610

Earlier quoted context omitted.

And did you use your 25 years of Java experience making $80k fixing waterfalls? lol

Without bad intent, I am not sure I am even able to make sense of your sentence. Honestly it sounds as if you fed my comment into an AI tool and asked for a reply. Here is a tip for the former junior dev turned nascent GenAI-Vibecoding-manager - if you want to attack someone's credibility, especially that of an Internet stranger you are desperately trying to prove wrong, try to use something they said themselves, not…

[flagged]

Re: 4o Image Generation

#628

Post by sayamqazi »

wegfawefgawefg wrote: post #588
vlovich123 wrote: post #549

Earlier quoted context omitted.

> Airplanes dont have wings like birds but can fly. and in some ways are superior to birds. (some ways not) I think you're saying exactly what I'm saying. Human brains work differently from LLMs and the OP comment that started this thread is claiming that they work very similarly. In some ways they do but there's very clear differences and while clarifying examples in the training set can improve human understanding…

sure. to be honest i dont really care if they work the same or not. I just like that they do work and find it interesting. i dont even think peoples brains work the same as eachother. half of people cant even visually imagine an apple. Neural networks seem to notice and remember very small details, as if they have access to signals from early layers. Humans often miss the minor details. Theres probably a lot more sig…

> half of people cant even visually imagine an apple.

What is the evidence for this? We are just taking people's word for it?

Re: 4o Image Generation

#629

Post by wegfawefgawefg »

sayamqazi wrote: post #628
wegfawefgawefg wrote: post #588

Earlier quoted context omitted.

sure. to be honest i dont really care if they work the same or not. I just like that they do work and find it interesting. i dont even think peoples brains work the same as eachother. half of people cant even visually imagine an apple. Neural networks seem to notice and remember very small details, as if they have access to signals from early layers. Humans often miss the minor details. Theres probably a lot more sig…

> half of people cant even visually imagine an apple. What is the evidence for this? We are just taking people's word for it?

youre one of todays lucky few. about to have your mind blown. look this one up.
Post reply on HN