Hacker News Reader

Comment by danielvaughn | Hacker News Reader

What's interesting to me is how AI-generated images (not just DALLE but also Midjourney and others) have a specific look and feel. It's typically characterized by high contrast and high saturation. Anyone know why that style is more likely to be the output?

View on HN

When you ask users the select the best picture, they will select the one with the high contrast and saturation.

It seems to me that I can get most any look and feel I want. Most of the ones other people post seem to have a very different look and feel to my own.

That said, you can sort of get a default look and feel if you just give a short prompt, and then it will tend toward the ones that are favored by RLHF. I prefer very long prompts.... as long as it will allow.

So if you do "a cool treehouse" you'll get sort of the default look. It will be very different if you say "treehouse, naturally occurring, in an old beautiful tree with branches that are low and spread widely and have lots of character and hanging moss and thick bark and curvy roots and mushrooms on a rocky outcropping from a mountainside. photograph, golden hour, sun through trees, damp from rain. Treehouse is part of tree, with fractal forms and live shaped wood and stone and stained glass and glowiness. art nouveau, gorgeous colors and fantasy design"

It's funny that the same people who complain that AI is "cheating" and uncreative, often are the ones who go to so little effort to get good results. It's not like it takes any arcane knowledge to get good images, but if you can use some imagination and string a lot of descriptive words together you can get so much better results.

One thing i found neat about ChatGPT DALLE-3 is that it's usually generated via prompts like yours. "a cool treehouse" actually used "Oil painting of an old-fashioned treehouse in a serene autumn setting. The tree's golden leaves provide a canopy for the cozy wooden hut with a thatched roof. A ladder leans against the tree, and children play below, collecting fallen leaves." as the prompt.

Furthermore, i found it's very easy to tweak the general design. A cool image, copy the prompt, tell it to make it more X with Y and Z, and you start producing a really neat prompt.

So far as someone with a lack of mind-image but who enjoys creating computer graphics (3d, animation mostly) it's proven as a really neat test bed. Hallucinations are almost a feature in this to me, granted these aren't strictly that - just saying i find it's RNG flavor over my prompt is really nice for exploring.

Sidenote, i entered your text - looks great!

"i entered your text - looks great!"

Yours does too. Very different feel than mine, but beautiful.

It’s just lack of aesthetic talent, while competent images they’re very tacky looking. It’s not fundamental to AI because if you know what you’re doing you can push it to interesting aesthetics. Although Midjourney does some things behind the scenes to push everything towards that look you’re talking about.

Midjourney has been collecting human preference data for about a year now - every time you generate an image there and click on the one you want to enlarge you're providing a signal as to which image a human being preferred.

So my hunch is that humans prefer high contrast and high saturation images!

Reminds me of The New Coke, a flavor designed by having people provide very short term feedback on different flavor combinations.

Most feedback processes for generative models are based on asking the user to draw immediate sentiments rather than having them provide deeper art and style critiques

Try photography terms, as they're less likely to have uploaded HDR'd content under postings that include the technical specs of what they shot with, e.g. "24mm f/1.2" or "70mm f/2.8" or "D820" and such.

50mm (optimally with "Nikon" or "Canon" or similar) or 35mm will probably get you the most "natural" looking FoV. (Adding "lower" # fstops will get you a lot more depth-of-field/bokeh.)

The old adage is "f/8 and be there" so f/8 might get the most natural images if you want to specify an fstop(?)

This is where things like img2img in StableDiffusion really came in handy as you could simply apply an entire prompt like a photoshop filter..

The same reason Apple started the fake "out of focus background" on non DSLR phone cameras. It makes people feel like its better

There's a lot of variation between monitors so high contrast/saturation is far more likely to look the same across monitors. Digital art in general has moved towards this style.

I've got two LG 27" 4k monitors, same model number but produced several years apart, and while one monitor can easily show light grays like #EEE, it just looks white in the other.

The CFG is a trick used to trade variety for apparent quality, applied to images it causes high saturation and contrast. It's used because the training data set is very noisy.

The paper: https://arxiv.org/abs/2207.12598, of course using CFG change the sample distribution from the training distribution giving it that specific look.

I think people just like it, so any RLHF like process will encourage it.

A lot of it is driven by underlying params - if you decrease style param a bunch in Midjourney, some of the high contrast & high saturation affinity goes away.

Only if you don't prompt more precisely.

It's likely from the RLHF

Basically on A/B tests, humans tend to prefer more saturated, "punchy" images. Which is also why iPhones tend to do the same thing.

For artificial images people also seem to prefer stylized "dramatic" styles as well.

And the model was finetuned to match.

Reply on news.ycombinator.com