I'm just going to ignore the trash talk that some people throw at others, and move on.

There is basically a prompt syntax for stable diffusion.

- Type of shot (portrait, full body, from behind, macro, etc.)
- Description of the subject
- Possibly control of the depth of field by means of aperture specification or e.g. "bokeh", "shallow depth of field". Mostly with f1.8 or f2.0. Focal length of objects can also be specified, but in my opinion it has no great influence, rather directly control the portrait type such as Extreme Close Up, Close Up, Medium Shot, Full Shot, etc.
- Additional information e.g. on foreground, background, colors, (rarely also composition, image harmony)
- For portraits, additional information on the facial expressions / gestures of the person (smiling, laughing, sad, angry, etc.)

Weighting works via bracketing.
Square brackets [] reduce the weighting, round brackets () increase the weighting.

Commas and periods play a role in separating information and content. Commas are light and periods are strong separators. However, this is actually very similar to normal punctuation when writing texts.
The same applies to the negative prompts.

With LORAS (fine adjustment models), additional weighting can be applied, e.g. with<lora:sdxl_photorealistic_slider_v1-0:1> the number after :

In addition, the images can be further fine-tuned via the number of steps (iterations) and the guidance scale.
With higher settings, the images generally become more detailed, but the prompts must be formulated very precisely.

Basically, I don't think much of controlling the type of image in the standard model by specifying "photorealistic", "artistic", "surreal", etc.
Instead, you should use this in combination with specially optimized models (see https://civitai.com/models).

I believe that open source Stable Diffusion can beat Midjourney in the long term because of the huge developer community and extreme flexibility.
But it's currently really pain in the ass to find right settings, which deliver good results.

I'm going to print that! Your answer led me to search for the correct question, and I found an additional FAQ. Thank You

Syntax for Stable Diffusion  https://www.generativelabs.co/post/prompt-syntax-for-stable-diffusion-faq

I see that the parts I was looking at are not f stops, but numerical weights. Also there's a 75 token limit. So some of the terribly long prompts I've seen could be truncated? Interesting.

I'm using DALL-E and SDXL 1.0

Accept AI images instead, much more promising..

No kidding.

I guess I can go delete the forum, which Alamy removed any post or discussion of Stockimo or mentioning that there was someplace for people who were trying to be active, could interact.

Oh I got the email: "It's with heavy hearts that unfortunately we�ve made the difficult decision to retire the Stockimo app


Over the last few months, we�ve been reviewing the performance of the app and have ultimately decided to move in a different direction.

Although we will continue supporting existing Stockimo contributors like yourself.

You�ll still be able to manage your uploaded photos and receive/review any sales, but we will NOT be accepting any new photo uploads from now on."

Does that mean, just the app? Crazy.

"Finally, my Administration will help develop effective labeling and content provenance mechanisms, so that Americans are able to determine when content is generated using AI and when it is not."
https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/

Honest in labeling, could be a good thing as fake AI, creative, imaginary images won't be used as supposedly real.

.

Quote from: SuperPhoto on January 27, 2024, 08:06

..

"Regulations" generally speaking only affect the little guy - and are used for anti-competitive behaviour.    ...

And lol - the only reason "ai" tools have difficulty with hands - is because they didn't have massive amounts of images to steal it from. People don't generally take pictures of their hands or feet and post those to social media, or stock media accounts. So it's kind of hard for them to steal it when the number is limited.

right, why does the little guy need regulations to, provide them with transportation safety, 40 hr work week, ensuring drugs are safe, clean air, clean water, etc etc lets go back to Dickensian times

hoist by your own petard winner this week:
https://petapixel.com/2024/01/24/trump-shares-ai-image-of-himself-praying-with-six-fingers/

as far as the hands problem, there are certainly enough hands & arms in the world image base, but the problem more likely lies in the fact that most pictures with people don't have hands, arms, etc as tags, so there's less to train by.

The problem with hands is, not all hands are flat out, four fingers spread, and a thumb. Some are holding things, some are pointing, some are shaking hands or holding hands. At that point and add variations of how many angles or fingers showing, the AI doesn't understand what a hand consists of, since there's no way to know the exact situation, and how many fingers are supposed to be there? We get the monsters.

b) The courts may "decide". If they decide to say that theft is not theft, it does not mean its not theft. One does not need a 3rd party "authority" to decide if something is wrong, when they know it is in fact wrong.

If you say so, but the courts make the laws.  I won't even try to say, what's right and wrong, different countries, different religions, different values. I think something like the Ten Commandments is pretty straight forward, what's right. (except for all those shalt nots, which could have been more positive instead of so negative?)

What I'm just pointing out is, if the courts say it's fair use, it's not stealing. If you call fair use, stealing, that's your choice, but you or I don't make laws or enforce them.

Please read this?

https://www.findlaw.com/smallbusiness/intellectual-property/fair-use-law.html

I'm just going to pick and choose, which is not in depth, but in the section where they explain, the four factors that the courts look at:

"The more transformative a new work, the more likely a court will consider it fair use."
"Courts consider how much material was copied and if the copied material is a central part of the original work."

If you use assorted learned pixels from 5 billion images, one might have to agree, it's transformative? And if you use 5 billion images to train the AI, how could the new image, be derived from the central part of one single image?

Reading a copyrighted book is not stealing, but we can learn from that and come up with new ideas of our own. Machine learning is doing just that, by viewing images. It is not "stealing" images, it's just looking at them.

And if the courts decide it's stealing and not fair use, I'm fine with that. I'm only looking at the law and what it is, instead of what I'd like it to be.

Quote from: Uncle Pete on January 26, 2024, 13:38

designed to illustrate articles on current events and newsworthy topics

So it can be news related. Yes, it will be interesting to hear what Matt has to say.

By the way, I like the cow avatar, Pete.

Thanks... Green and Gold Packer cow, with cheese and beer. It's all Wisconsin.

Regarding illustrative editorial requirements for photos on Adobestock. Adobestock does not accept Illustrative editorial images with recognizable people in them.
I understand that if you took an illustrative editorial photo of say the exterior of a store that it would not be accepted if someone happened to be walking by and was in the photo.
I have always assumed that you are not allowed to "photoshop" such person out of the photo and then submit it.
However, in reading the editorial guidelines, it doesn't specifically say that.

So, my question is: are you allowed to remove people from a photo and then submit as illustrative editorial? I am pretty sure the answer is no, but thought I'd ask to be sure.

I know Mat reads here and especially this subject. Maybe he's waiting for an answer, or how to answer, from Adobe?

In an attempt to fight fraud, unsuspecting parents who were receiving child support from the state were unjustly targeted (via racial profiling) and accused of fraud. Even though they did nothing wrong, they had to pay back the full amount of child support they'd received over the years, causing huge debts. When they tried contacting the tax authorities, they always received canned responses and accusations and nobody listened to their cries for help. It was like talking to a brick wall. Very kafkaesque.

Prisons all over are full of innocent people, just ask any of the inmates?

Is the legal system in the Netherlands that messed up, that they couldn't file a legal appeal through the courts? Or that they couldn't find an attorney to act on their behalf. Do you know these people, personally? Ask them why they only hammered on breaking a brick wall, instead of going around it.

Microstock is different. No legal system, only the agency as authority, and yes, one department, probably doesn't communicate with the others.

But someone being penalized for success is quite the slap at the artist and OP.

Quote from: Uncle Pete on January 25, 2024, 12:05

Doesn't work for me, it shows the Featured Collections.

Scroll down, it's below Featured Collections.  It's way easier than manually manipulating URL. Firefox, Mac

I bookmarked the link years ago, so easier than going to AS and clicking contributor account. 

The OP said his were not showing on that page, which you are correct, shows the same as the link, just below the featured collections.

best for submitting:  ...DT only allows 1 image at a time

Please explain? One image at a time? They have copy keywords and a kind of batch thing. I don't use it, because mostly nothing I do is the same, but there's some kind of populate from previous image. Or did you mean something else.

DT is the easiest for me to upload, especially vectors, where I upload the JPG and then the "additional". They match them, include the data and, it's done. Automatic categories. Worst part of DT is, yes I get accepted, no I don't sell much.

what is the best agencies for vectors ?

Good question, I wonder if I'm missing something by only working SS, AS, DT? IS I get some DLs but uploading is such torture that I stopped. I wonder if there is another that might be better for vectors than for photos?

Quote from: cobalt on January 25, 2024, 05:49

the link is not working

I think Rob meant this link, but accidentally made an error with the link formatting: https://www.facebook.com/fotoproduzent/posts/pfbid02zqdMHyC661xGNrZeQqxZteatYjfMbFcoRgxLFexagFMJW9rvjdCdVFBAVQdnvAel

That one works:  https://www.facebook.com/fotoproduzent/posts/pfbid02zqdMHyC661xGNrZeQqxZteatYjfMbFcoRgxLFexagFMJW9rvjdCdVFBAVQdnvAel

My top ten agencies? I only have one. The rest don't make enough to be ranked.

Adobe Stock

I guess what I'm looking for is one more website to constantly reload the contributor summary page on.  Seems to me it's pretty clearly AS in first place, istock in a distant 2nd and SS is trying to win the race for last place (for photo and a little video).  If I'm going to waste time submitting to SS maybe I can waste a little more at dreamstime or depositphoto or something?  Seems dreamstime is the consensus 4th least worst?  Correct me if I'm wrong.

Nope you're not wrong. The problem comes down to, what are your images and how many? Then how much time will you have to spend to drop those images into a black hole?  Yes, DT is probably the next best after the three, and maybe DP is next after that.

If there was a 4th I'd want to know that too.

My point of view is, there are only 3 right now. Yes I have a DT account, I'm trying to make it to $100 so I can close it. BUT... I have a friend or two that make payout there, every few months. I think there are others here who say they like the returns from DP. And there are people who have dropped SS when the reset and 10c commissions came in.

In the end, you'll have to try and see or depending on what your images are, see what your personal results are. Mine say, AS, SS and IS. All the rest that I hang on to, are dying or dead.

what u smoking bro? I have images waiting from 29 days ago...

I have a variety, oldest is "last month" newest is 6 days ago. But in between that, some have been accepted and are live. 19 days, 15 days, 12 days, 6 days. I wouldn't call that very quick. And I promise on a good day, has never been faster that SSTK.