It's easy to ask all community than actually use a tool called search for it first isn't it? eheheh
i give you a clue. use it.

Topic: ChatGPT AI Actually creates good keywords

Quote from: Evaristo tenscadisto on October 23, 2023, 03:49

It's easy to ask all community than actually use a tool called search for it first isn't it?

Have you tried using your own tip? Did you find in search any results relevant to microstocks?

I'm asking about a personal experience, relevant to microstocks. Not "how to use ChatGPT for photography".

Probably i was not clear since i was trying to make a joke and respond to you at same time what you are asking. But I can guarantee that the joke is not on you but on the situation: it involves AI and the type of responses that Chatgpt sometimes makes if you ask him to play a role.

This time I'm going to try to keep more unboxed: Yes. ChatGPT AI actually creates good keywords. My advice/tip is use it!

However i did try to ask Chatgpt, without role playing,  your question.
Here is the result:

"As of my last knowledge update in September 2021, ChatGPT, like other AI models, was not typically used in the context of microstock photography or similar industries. Microstock refers to the sale of stock photos, illustrations, and other digital media through online platforms, often for a lower cost than traditional stock photography.

However, AI and machine learning technologies have been applied in various ways in the field of photography. This might include image recognition, automated tagging, and content recommendation systems. These technologies can help streamline the process of managing and searching for stock images, but it's typically not the AI model like ChatGPT that is directly involved in these tasks.

If there have been any developments or specific applications of ChatGPT or similar AI models in the microstock industry since my last update, I would not have that information. I recommend checking with the latest industry news or consulting with professionals in the field for the most up-to-date information on AI usage in microstock and related sectors."






 

What query did you use to get "good" keywords, and how many were you able to get at a time?

I've found generally speaking the results have been inconsistent, and limited. How were you able to get good results?

Quote from: ribtoks on October 23, 2023, 04:03

Quote from: Evaristo tenscadisto on October 23, 2023, 03:49

It's easy to ask all community than actually use a tool called search for it first isn't it?

Have you tried using your own tip? Did you find in search any results relevant to microstocks?

I'm asking about a personal experience, relevant to microstocks. Not "how to use ChatGPT for photography".

Probably i was not clear since i was trying to make a joke and respond to you at same time what you are asking. But I can guarantee that the joke is not on you but on the situation: it involves AI and the type of responses that Chatgpt sometimes makes if you ask him to play a role.

This time I'm going to try to keep more unboxed: Yes. ChatGPT AI actually creates good keywords. My advice/tip is use it!

However i did try to ask Chatgpt, without role playing,  your question.
Here is the result:

"As of my last knowledge update in September 2021, ChatGPT, like other AI models, was not typically used in the context of microstock photography or similar industries. Microstock refers to the sale of stock photos, illustrations, and other digital media through online platforms, often for a lower cost than traditional stock photography.

However, AI and machine learning technologies have been applied in various ways in the field of photography. This might include image recognition, automated tagging, and content recommendation systems. These technologies can help streamline the process of managing and searching for stock images, but it's typically not the AI model like ChatGPT that is directly involved in these tasks.

If there have been any developments or specific applications of ChatGPT or similar AI models in the microstock industry since my last update, I would not have that information. I recommend checking with the latest industry news or consulting with professionals in the field for the most up-to-date information on AI usage in microstock and related sectors."

I will say this, this ChatGPT may offer no more than 10 tags on the topic, but it does not offer anything new.

Quote from: SuperPhoto on October 23, 2023, 05:14

I have used it, and overall it's very difficult to get good consistent results - such that I've found doing it the "old school" way is much better. I now usually use from time to time just to "complement" my original workflow (but even then, debating its usefulness because of the extra time required to sort through crap).
...
As of yet - for myself - I have not yet figured out an efficient way of getting consistent, useable information.

Thank you for your detailed response! This was largely my experience too. But still I was interested if anybody managed to make any use of it.

By the way, did you use free version (GPT 3.5) or Pro (GPT-4)?

Used the free version (3.5).

Was considering trying/using the paid version, but haven't yet. So that was my experience with the "free" version.

...so anything that is slightly not in line with the "approved mainstream t.v./govt narrative"...

Okay, sounds bad

"climate change ... designed to push a carbon tax to tax people more to try to take their wealth and restrict their movement"...
...we're spraying the sky with chemicals "for your safety & protection"'...
...the shots actually do cause cancer/infertility/etc, which was by design...
"mathsks" actually did have poisonson them to "make ppl sick"...

 
LOL

Honestly the forum is getting full of it. Hard to keep track of who's a "their among us guy", who's a Q guy, who�s a chem trail guy, a white replacement theory guy, a �the neo-Marxists are out to get us� guy,  a far right ethnonationalist or libertarian, an incel/ misogynist on and on.

Guess it's the destiny of all online forums eventually. The people who are put off by the madness slowly just back out the room. It just isn�t a pleasant place to hang out.

Honestly the forum is getting full of it. Hard to keep track of who's a "their among us guy", who's a Q guy, who�s a chem trail guy, a white replacement theory guy, a �the neo-Marxists are out to get us� guy,  a far right ethnonationalist or libertarian, an incel/ misogynist on and on.

Guess it's the destiny of all online forums eventually. The people who are put off by the madness slowly just back out the room. It just isn�t a pleasant place to hang out.

Haha, does that make you  the 'believes 100% everything on t.v. and totally obedient to what the newspapers say' guy? BTW - probably if you are finding a lot of people are trying to get through to you... then it might be worth looking into/using your own thinking to look at its merit... There is most certainly a difference between a "conspiracy theory", and just plain old "conspiracy"... (as well as the label 'conspiracy theory' is a tactic to make someone short-circuit their thinking, and just automatically try to 'dismiss' something without looking at the merit of it). But, I do understand with years of schooling for most people - being taught to be 'obedient to authority', and that 'authority is right without question', it can be hard to break out of that thinking to see things as they really are... Good luck.

I've never used it for keywording, but it has been a great sparring partner for me in learning Illustrator scripting, which I have used to make algorithmic-based stock vectors.

I have found for basic scripting it can be good... but if you try and get anything a little more sophisicated (i.e., say sophisticated patterns, etc) it does a really poor job.

Quote from: Justanotherphotographer on October 23, 2023, 07:08

Honestly the forum is getting full of it. Hard to keep track of who's a "their among us guy", who's a Q guy, who�s a chem trail guy, a white replacement theory guy, a �the neo-Marxists are out to get us� guy,  a far right ethnonationalist or libertarian, an incel/ misogynist on and on.

Guess it's the destiny of all online forums eventually. The people who are put off by the madness slowly just back out the room. It just isn�t a pleasant place to hang out.

Haha, does that make you  the 'believes 100% everything on t.v. and totally obedient to what the newspapers say' guy? BTW - probably if you are finding a lot of people are trying to get through to you... then it might be worth looking into/using your own thinking to look at its merit... There is most certainly a difference between a "conspiracy theory", and just plain old "conspiracy"... (as well as the label 'conspiracy theory' is a tactic to make someone short-circuit their thinking, and just automatically try to 'dismiss' something without looking at the merit of it). But, I do understand with years of schooling for most people - being taught to be 'obedient to authority', and that 'authority is right without question', it can be hard to break out of that thinking to see things as they really are... Good luck.

"    Historical Exploitation: Throughout history, colonial powers, often led by European males, have oppressed and exploited indigenous populations in various parts of the world. This exploitation included forced labor, land theft, and cultural suppression.

    Slavery: The transatlantic slave trade, largely driven by European powers, resulted in the forced enslavement of millions of Africans. This brutal system of oppression led to centuries of suffering and continued racial disparities.

    Racial Discrimination: Discriminatory practices, including segregation and institutional racism, have marginalized racial and ethnic minorities in many societies. This discrimination has resulted in disparities in education, employment, housing, and criminal justice, among other areas.

    Stereotyping and Prejudice: Minorities often face stereotypes and prejudice perpetuated by the dominant culture, which can lead to bias in various aspects of life, including employment, education, and social interactions.

    Cultural Appropriation: The appropriation of cultural elements from minority groups by those in power can lead to the erasure of cultural identity and perpetuate harmful stereotypes.

    Unequal Access to Resources: Systemic disparities in access to resources, including economic opportunities, healthcare, and quality education, can disproportionately affect minority communities.

    Violence and Hate Crimes: Hate crimes and racially motivated violence can target minority individuals and communities, causing physical harm and psychological trauma.

    Disproportionate Incarceration: Racial and ethnic minorities, particularly in the United States, are often overrepresented in the criminal justice system, facing harsher sentences and unfair treatment.

    Political Disenfranchisement: Minority groups have historically faced barriers to political participation, including voting restrictions and gerrymandering, which can limit their ability to have a voice in decision-making."

Wow those evil WASPs?

Back on topic, it would be nice if AI could look at an image and say, here are keywords, for us. On the other hand, isn't that what keyword suggestions already do on Adobe, SS and DT, for example?

Or is there something in "please generate good keywords for an image of sliced vegetables" and then we use the results to add to are already thoughtful keywords?

Hey, not bad? I see some I could use.

    Sliced vegetables
    Fresh produce
    Culinary preparation
    Food preparation
    Chopped veggies
    Colorful ingredients
    Healthy cooking
    Salad ingredients
    Kitchen ingredients
    Food diversity
    Vibrant colors
    Nutrient-rich
    Cooking ingredients
    Culinary art
    Ingredient diversity
    Sliced carrots
    Chopped bell peppers
    Vegetable medley
    Knife skills
    Nutritious meal prep

I'm happy the Chat GPT 3.5 supports "Ingredient Diversity" for vegetables. 

Back on topic, it would be nice if AI could look at an image and say, here are keywords, for us. On the other hand, isn't that what keyword suggestions already do on Adobe, SS and DT, for example?

Midjourney has a nice option, to /describe an image for reprompting. You have to upload an image and you can obtain 4 different descriptions, descriptions that often contain useful specific terms and look good for chatGPT when you ask for keywording.
Unfortunately, the time to:
1- produce a good image
2- upload to midjourney and ask for descriptions
3- check descriptions and use them to obtain good keyword list from chatGPT
4- check the keywording ChatGPT job to find out how accurate is it, what to delete, and the right keywords order in alphabetical order

... is by far longer than simply do the usual semi manual keywording

I've never used it for keywording, but it has been a great sparring partner for me in learning Illustrator scripting, which I have used to make algorithmic-based stock vectors.

This is really cool! Can you share any "hello world" example what exactly did you do? All things aside (e.g. microstock, money etc.), it sounds just like quite a unique and fun thing to do! (you could send a DM, if you don't want to share publicly)

P.S. I don't mean the code of any of your algorithm-based picture, but more of things that surround that, e.g. how did you create an illustrator plugin/script, what did you ask ChatGPT to help with exactly etc.

Haha, does that make you  the 'believes 100% everything on t.v. and totally obedient to what the newspapers say' guy?...

LOL, I�m just the maybe this stuff doesn't need to be injected into every other thread on a microstock forum guy. Another person has just posted, unprompted and completely without irony, a wall of text about how they are racially oppressed, (as a WASP) then has the gall to post �back on topic�. A repeated refrain from them. If you don�t want threads derailed and to stay on topic why post that nonsense in the first place.

Is the thinking really it�s fine for one person to post massive rants about how down trodden they are as a white male, or for that matter how Covid is a conspiracy, but not fine for anyone to respond?

By posting that stuff you are either asking for a thread to be derailed or, alternatively, that you should get to post divisive opinions and no one else should get to respond in kind. In Pete�s case often that you should get to post a massive essay reading like a Tucker Carlson section then  say �back on topic� like you are the adult in the room.

Pete, you even started a thread elsewhere where this nonsense could be posted, why not keep it in the �Off Topic� section instead of posting about your aggrievement and repression all over the place. It�s like you�re obsessed with this stuff?

Yeah, all this off topic talk is making it much harder for you to farm us for content for your blog

Yeah, all this off topic talk is making it much harder for you to farm us for content for your blog

Nah, I'm just disappointed to experience Godwin's law (https://en.wikipedia.org/wiki/Godwin%27s_law) again and again on the internet, even in threads that I start!

P.S. You are free to compile responses from this thread to a blogpost of your own, if you think this is how it works and that it's so easy

No one called anyone a N**i though. It would be more accurate to coin a term for an "SJW" or "Woke" or "Neo-Marxist" Law. Far more likely someone's gonna get called that nowadays.

Anyway, I have nothing against Pete being an SJW or Neo-Marxist with politics of aggrievement for his racial group. Or that he is �woke� to the systemic policies of repression of the white man by the new world order. I make no judgement as to the validity of his aggrievement. I can't speak to the suffering he has suffered under the heel of whoever (though I can't remember anyone bringing up his race on this forum other than him? that could be a lapse in my memory?). But does it have to be forced down everyone�s throat? Can't we live our lives without these woke SJW people and their agendas?

I also don�t care if someone else believes in chem trails, that covid and climate change are a lie. Whatever, it�s just that there�s an off topic section especially for this stuff.

Quote from: Justanotherphotographer on October 24, 2023, 06:06

No one called anyone a N**i though. It would be more accurate to coin a term for an "SJW" or "Woke" or "Neo-Marxist" Law. Far more likely someone's gonna get called that nowadays.

Anyway, I have nothing against Pete being an SJW or Neo-Marxist with politics of aggrievement for his racial group. Or that he is �woke� to the systemic policies of repression of the white man by the new world order. I make no judgement as to the validity of his aggrievement. I can't speak to the suffering he has suffered under the heel of whoever (though I can't remember anyone bringing up his race on this forum other than him? that could be a lapse in my memory?). But does it have to be forced down everyone�s throat? Can't we live our lives without these woke SJW people and their agendas?

I also don�t care if someone else believes in chem trails, that covid and climate change are a lie. Whatever, it�s just that there�s an off topic section especially for this stuff.

Ah, so you are just trolling? Okay, well - if you want to get back on topic, do so. You are the one who derailed it. Was it fun?

Get back on topic then.

Or... are you just demonstrating the usage of ChatGPT for producing a response, and showing how useless it can be/the gibberish it can produce ? If so, well xribtoks - there's a demonstration for you!

(Haha, just kidding photographer guy - if it actually was a legit response - who knows - you could have used chatgpt for that response - but, if you were just trolling lol, okay - get back on topic then).

Yes, I guess sometimes a bit of humor or sarcasm doesn't come through on a forum. My post was in support of your experience with Chat GPT have a leaning towards revisionist history, meaning, everything now is universal and multicultural and the old days were all controlled by horrible, nasty, mean, invading, racists. My favorite part was "Cultural Appropriation: The appropriation of cultural elements from minority groups by those in power can lead to the erasure of cultural identity and perpetuate harmful stereotypes."

And no I didn't really bring up my heritage? I'm an American. I was born here and so were my parents, But if anyone wants to actually understand, My Grandfather came over on a boat from Italy to escape persecution and poverty, to a better place. He wasn't a mobster, or associated with the Black Hand. And back then, Italians were persecuted in the US, got the low pay and crappy jobs.

"Anti-Italianism arose among some Americans as an effect of the large-scale immigration of Italians to the United States during the late nineteenth and early twentieth centuries. The majority of Italian immigrants to the United States arrived in waves in the early twentieth century, many of them from agrarian backgrounds. Nearly all the Italian immigrants were Roman Catholic, as opposed to the nation's Protestant majority. Because the immigrants often lacked formal education and competed with earlier immigrants for lower-paying jobs and housing, significant hostility developed toward them."

Yeah, get over it. They adapted to the American life and culture and assimilated into society. Oh yeah, they all lived in an Italian neighborhood in Chicago and still spoke Italian at home.  I have cousins that still do. It's a choice.

The other Grandfather escaped from Transylvania, under similar circumstances. Religious oppression, poverty and social stigma, as the Transylvanian Saxons were the lowest of the low.

So next time you make a joke about some gangster in the Mafia, or Italians that are all mobbed up or Dracula being a vampire from Transylvania, I want you to be politically correct and stop. No harmful stereotypes.  You wouldn't want to culturally generalize, stigmatize or appropriate either of those. If you continue my Cousin Mario will come with a baseball bat, and give you a lecture on how to be polite around WOPs and guinea hens.  Or the other kind of bat, will bite you on the neck, and you'll be forever sleeping in a coffin and only able to come out at night. (I never did work out how that garlic thing could be overcome, as just about everything I cook has garlic, oregano and basil in it. But that's supposed to keep away vampires?  )

Quote from: Uncle Pete on October 23, 2023, 12:44

Back on topic, it would be nice if AI could look at an image and say, here are keywords, for us. On the other hand, isn't that what keyword suggestions already do on Adobe, SS and DT, for example?

Midjourney has a nice option, to /describe an image for reprompting. You have to upload an image and you can obtain 4 different descriptions, descriptions that often contain useful specific terms and look good for chatGPT when you ask for keywording.
Unfortunately, the time to:
1- produce a good image
2- upload to midjourney and ask for descriptions
3- check descriptions and use them to obtain good keyword list from chatGPT
4- check the keywording ChatGPT job to find out how accurate is it, what to delete, and the right keywords order in alphabetical order

... is by far longer than simply do the usual semi manual keywording

Is the free Midjourney able to do this?

And just to be clear. Like others, I have standard sets of keywords, I attach them to images from groups, the easiest is my Racing keywords, which get dropped in. But I wouldn't be against at least looking to see what AI has to say. Maybe I could get some more. And last I checked, I removed some that were kind of distant and I had never seen anyone, anywhere, except us stock folks, use as a search or keyword. Why load up images with fluff that serves no purpose?

Does anyone here call an auto race a "motorshow" I'm cutting that one.

Maybe I could try to work around the obvious and use Chat Gpt and ask, "Please name some words that would be associated with motorsports and motor racing"

Glad you brought this up. Something for a snow day.

What query did you use to get "good" keywords, and how many were you able to get at a time?

I've found generally speaking the results have been inconsistent, and limited. How were you able to get good results?

Quote from: Evaristo tenscadisto on October 23, 2023, 05:03

Quote from: ribtoks on October 23, 2023, 04:03

Quote from: Evaristo tenscadisto on October 23, 2023, 03:49

It's easy to ask all community than actually use a tool called search for it first isn't it?

Have you tried using your own tip? Did you find in search any results relevant to microstocks?

I'm asking about a personal experience, relevant to microstocks. Not "how to use ChatGPT for photography".

Probably i was not clear since i was trying to make a joke and respond to you at same time what you are asking. But I can guarantee that the joke is not on you but on the situation: it involves AI and the type of responses that Chatgpt sometimes makes if you ask him to play a role.

This time I'm going to try to keep more unboxed: Yes. ChatGPT AI actually creates good keywords. My advice/tip is use it!

However i did try to ask Chatgpt, without role playing,  your question.
Here is the result:

"As of my last knowledge update in September 2021, ChatGPT, like other AI models, was not typically used in the context of microstock photography or similar industries. Microstock refers to the sale of stock photos, illustrations, and other digital media through online platforms, often for a lower cost than traditional stock photography.

However, AI and machine learning technologies have been applied in various ways in the field of photography. This might include image recognition, automated tagging, and content recommendation systems. These technologies can help streamline the process of managing and searching for stock images, but it's typically not the AI model like ChatGPT that is directly involved in these tasks.

If there have been any developments or specific applications of ChatGPT or similar AI models in the microstock industry since my last update, I would not have that information. I recommend checking with the latest industry news or consulting with professionals in the field for the most up-to-date information on AI usage in microstock and related sectors."

I'm sorry to hear that you're not having a good experience with chat gpt but for me its been great. I have been using for almost a year now and I've been getting good results.  Sometimes I change 4 or 5 out of 40. 

It's no big secret sauce but to keep it simple I made a template where I give the instructions. it improves my title and gives me 40 keywords, not just descriptive ones, but conceptual and emotional ones where the clip can also be used. Therefore, not only can be used for documentary but also for storytelling and other point of views. In the instructions I set to do it for 50-100 or different clips at once.

In just over ten seconds I have all the work done by chatgpt where I only analyze and rectify those that I think are exaggerated for the image/video. In less than 15 minutes I made close to 250 different clips where I practically only lost time copy pasting from one side (chat gpt) to the other (website). Probably 15 minutes is overkill... maybe 5-7 minutes will be more realistic.

With the old methods of self-generating in microstock website or using a keywording tools, i will waste more time. much more time! I really recommend it especially for those which English is not the native language. They can write in their own language and have the work in perfect English with good and refined keywords.

Anyway if chatgpt doesn't work for you, then use other "guitar" that works best.

Quote from: Evaristo tenscadisto on October 26, 2023, 02:05

It's no big secret sauce but to keep it simple I made a template

This is interesting!
May I ask you more details about this? How a template can be adapted to many different images (or clips)?
Is it necessary to have the pay version or it can also be done with the free chatGPT?
Thanks

I use the free version. you don't need to pay.
The template is not about code or anything programmed. 

Although these days requesting tasks on chatgpt is a kind of programming the template that i created is just a set of tasks followed by considerations that I ask  chat GPT to take into account before performing the work. Then I add to that the list of 50-100 clips with the names I rendered. Clip names are simply basic descriptions of what the clip contains.

If you are going to create one keep in mind simple tasks with direct instructions. Don't use to much "Considerations" to perform the work. One for technical aspects and one for behavior/subjective aspects is enough. If you put more you start making entropy for chatgpt and you start getting redundancy in the feedback of your keywords. Less is more accurate if well defined at the begin.

In short the template structure is:
1 - Purpose of task
2 - Simple tasks with direct instructions (i only use 5 tasks)
3 - Considerations/work performance (How the work will be done by chatgpt in terms of Form and Content)
4 - List (list of names described. i use 50 or 100)

Hope it helps you. 
 

Hope it helps you. 

Thank you very much Evaristo for taking time to explain, really interesting!

I actually use ChatGPT currently to create prompts.
I usually get inspired by an interesting photo/image, write a brief description or some keywords of the content/topic and then rephrase it via ChatGPT.
This saves a lot of time and increases productivity.
Sometimes I even create new and better image compositions.

My perception is also that the more extensive the prompts, the more detailed images can be created with a higher prompt guidance scale in Stable Diffusion.
It's crap to use keywords like "higly detailed", "photorealistic", blablabla, etc. in prompts.

As Evaristo tenscadisto has already written, it makes sense to create templates with short, specific instructions.
Sometimes, for example, I only paste in 3 to 5 keywords and create a detailed prompt because ChatGPT knows from previous example descriptions and specifications how I would write the prompts myself.
As a power user, you could certainly try to automate the whole thing.

For example, I am currently trying to generate authentic culture-specific portraits.
The repetitive is sometimes even desired.

I would say this is still the lowest beginner level and can be extended.

I have given ChatGPT some examples of how the prompt should be constructed with a description of the person, culture and landscape in the background.

These are two following generated prompts and generated images in Stable Diffusion.
(raw images are from the first run and must either be adjusted in Stable Difussion by variations or manually retouched after upscaling).

Keywords:
african female, authentic dressed, backdrop sunset savannah

Prompt:
Detailed close up of an African woman adorned with a traditional headwrap, symbolizing humble beauty against the backdrop of an African savanna with a breathtaking sunset. The scene captures the essence of cultural richness and natural splendor, celebrating the woman's grace and connection to her heritage.

Image:
https://ibb.co/brF4dxz

Keywords:
spanish female, authentic dressed, dancing flamenco, backdrop madrid sunset

Prompt:
Detailed Close Up of an elegant, traditionally dressed Spanish woman passionately dancing Flamenco in front of madrid city against the backdrop of a breathtaking sunset. The scene captures the cultural richness and emotive beauty of Flamenco, blending the artistry of dance with the captivating hues of the evening sky.

Image:
https://ibb.co/ZYf4dhW

I currently need about 1 day to generate a certain set-up (approx. 30 to 40 images), sort them out manually and retouch them. That leaves me with about 10 to 15.
I try to generate daily up to 5 max 8 set-ups, so about approx. 50 to max. 120 final images a day. In the meantime I prepare the following prompts.
Currently it is still a very manual and multi tasking task.

But I currently earn a maximum of 10 USD / month with it.
I still have almost 800 images in review on Adobe ...

Does AI use stop words or ignore them? Does punctuation have any value? Why do I see prompts that have things like f11:10 on the end? What does that mean?

I'd like to know what actually matters, instead of just throwing out words and having some random seed start to make a image that's using some of that and ignoring other parts.

Anyone have a guide?

I've tried a number of times, just for an example, to get SDXL to make "A cow jumping over the Moon" Never one, it puts in a cow, on the moon, over a moon background, all kinds of things, but not a cow, above the moon or jumping over it. DALL-E does better but low quality output.

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

I had a look at the mage page.

In the terms of service it says that everything created is public domain.

Not sure if that is ideal.

I had a look at the mage page.

In the terms of service it says that everything created is public domain.

Not sure if that is ideal.

Everything created is public domain because AI can't be copyrighted. A machine is not a human mind.

Quote from: Andrej.S. on February 01, 2024, 18:45

@Uncle Pete
Interesting, I didn't know about the token length restriction in Stable Diffusion. Thanks for the info!
I'll try it out with longer, detailed prompts.

Which online provider do you use?
Or do you generate locally on your own machine?

I currently use Mage.Space. Can't complain. For 15 USD / month unlimited generation with private mode and many pre-installed models / loras.

The cool thing is that in combination with precise prompts and special models, you can maintain character continuity quite well without too much deviation.
So it has potential for image series. I haven't seen sofar Midjourney users using this, but maybe nobody thinks about it.

I'm currently trying to generate colorful portraits with African-American models:
https://ibb.co/PhRSgNN
https://ibb.co/PTkRFZZ

And also the current body positivity, female sexual wellness trend:

https://ibb.co/f1sWkQR
https://ibb.co/1J0f0mM

Oh and yoga in a more "raw" style:
https://ibb.co/nLPSjnw

yes, the hands and minor artifacts are still really annoying at the moment ...

But as I said, I see much more potential in Stable Diffusion than Midjourney to stand out through individualization from the huge same-looking crowd. The buyers will be soon so pissed off because of millions of same looking stuff.

Just a commentary on the 'body posivitity' thing. It's funny how the tell-a-vision & media has convinced people being "fat" and "obese" is a "good thing" when it's actually very, very bad, by inventing new words & garbage phrases like "body positivity", when it's actually very, very bad to be that disgustingly overweight and fat. Being fat & overweight is not something to be proud of. It means you need to stop eating the potato chips, refined sugar, canola oil products, get off the sofa, get some exercise, stop taking meds & injecting yourself with crap. It means you are sick, because you have a bunch of parasites & crap inside of you that your body can't get rid of.

Pretty messed up how "they" (the people who own/run the wef/imf/UN/etc) appear to have confused so many people.

If I uploaded a picture like that, I'd be tempted to call it "Fat slob pretending to be happy by being slothful". It's a much more accurate description.

Absolutely agree. I like your straight thinking and cynical manner. Yeah the sheep mass mob, which we call people (in German we would say "das gemeine Volk") are just trend runners. Of course, being overweight is no reason to be proud of a fat body.
After years of vomiting in order to be slim, women are now encouraged to guzzle and swallow themselves healthy.
And why the heck there is no body positivity for men? Or do you see any posters with happy curly men?

The same applies to sustainability, diversity, etc.  ... Things, which will vanish in 10 to 15 years and have no real inherent, long-lasting values, because they come from outside and not from the thinking of society itself.

Quote from: Year of the Dog on February 02, 2024, 07:04

Quote from: cobalt on February 01, 2024, 22:21

I had a look at the mage page.

In the terms of service it says that everything created is public domain.

Not sure if that is ideal.

Everything created is public domain because AI can't be copyrighted. A machine is not a human mind.

not true - the basic law hasn't changed so images are copyrighted by their creators (humans using tools) and there are cases i n several jurisdictions pending - even there, a decision on one US circuit does not apply to others. and EU and th er countries may have differing standards.

AI gen is not a 'machine' but software , the same as PS or others. the mind doesn't creates ideas, but it takes a tool/machine whether pencil or camera or computer to create the expression of that idea which is automatically grantved copyright

US copyright office disagrees with you, and contradicts itself, but China says they are protected. The EU doesn't know what they think. There's no unity to the union. When the supreme court decides we'll know.

Read this thread about it:
https://www.microstockgroup.com/general-stock-discussion/keywording-and-meta-data-gpt/

https://www.reddit.com/r/stockphotography/comments/17yd2ek/stock_photo_metadata_guru_gpt/?utm_source=share&utm_medium=web2x&context=3

in an often cited case, they rejected a copyright when the owner oddly claimed the ai was the creator and he owned the copyright as a 'work for hire' - the office rejected these arguments, which is not the same as saying an artist cannot copyright a work they CREATED using a computer program.

https://www.copyright.gov/ai/docs/us-cross-motion-for-summary-judgment.pdf

the core of the decision was the Office �will not register works produced by a machine or mere mechanical process that
operates randomly or automatically without any creative input or intervention from a human
author.�  and the applicant undermined his case by declaring the work was �created autonomously by machine.�. this judgment was different from declaring ai assisted works cannot be copywrote.

my emphasis shows the issue that  remains to be directly litigated or legislated.

Yes, but you are biased towards one side of the decision and cherry-picking your defense.    https://www.goodwinlaw.com/en/insights/publications/2023/08/insights-technology-aiml-thaler-v-copyright-register-gen-ai

So I'll do the same:

When an AI �receives solely a prompt from a human and produces complex written, visual, or musical works in response, � users do not exercise ultimate creative control� and the resulting work is not copyrightable

When a human selects, arranges, or modifies AI generated material in a sufficiently creative way, the work may be copyrightable

Yes, AI work can be copyrighted, if a human selects, arranges or modifies... Thaler could have said, he edited the output and then what?

Courts in the United States, Europe, Australia, and New Zealand, have all held that only natural persons can be considered inventors.

Yes there is investigation and there will be more cases, but I don't think this is going to be as simple as, someone typed the prompt, so it's something that can be protected. Will AI chat text and articles be protected, because someone asked for the subject?

My best argument in favor of our work being protected would be, I generate the prompt, and there's an image, which, after that, I the human, edited and altered into a new work, which therefore means I can protect my human creation, based upon an AI, public domain image. 

A �guiding human hand� is necessary for authorship, and I am that human.

yes, that's my biased opinion - your 'biased' is my ' intelligently informed opinion'

Thanks for understanding that I'm also intelligently informed and just came to a different biased conclusion. 

Two very simple points, as the laws are now.
1) Only natural persons can be considered inventors
2) AI creations from scraping the web for training, are fair use.

For #1 copyright will be rejected, no matter how intricate the whole creation process may be, or whether it can be detected as AI or not. #2 the current laws defend the right of the AI companies to look at images and use that to create training.

For myself, on the other side, I still say, we added the captions, or some human did. (unless someone uses AI keywording) The text, description and keywords, included in the images, is our work and individual writing, which can't be used for training as fair use. That's my data and I own the right to that. The keywords are being illegally appropriated.

Quote from: Uncle Pete on February 05, 2024, 15:23

Quote from: cascoly on February 04, 2024, 14:21

yes, that's my biased opinion - your 'biased' is my ' intelligently informed opinion'

Thanks for understanding that I'm also intelligently informed and just came to a different biased conclusion. 

Two very simple points, as the laws are now.
1) Only natural persons can be considered inventors
2) AI creations from scraping the web for training, are fair use.

For #1 copyright will be rejected, no matter how intricate the whole creation process may be, or whether it can be detected as AI or not. #2 the current laws defend the right of the AI companies to look at images and use that to create training.

For myself, on the other side, I still say, we added the captions, or some human did. (unless someone uses AI keywording) The text, description and keywords, included in the images, is our work and individual writing, which can't be used for training as fair use. That's my data and I own the right to that. The keywords are being illegally appropriated.

detail #1 - inventions can't be copyrighted but can be patented, so the question comes back to creation. i spent an afternoon earlier and my initial prompt didnt give what i asked for, but eventually used it to springboard & eventually ended up with 6 distinct series - it was not a simple prompt = final image; instead using a tool to refine whhat the tool gave me

 individual keywords can be copyrighted,. and since the training only uses individual words, there's no violation.

the main claim for fair use is that by scraping billions of images, no individual image is copied. instead each image is broken down to tokens and there's no way to reverse the process to identify where a token came from

The word inventor applies to art, music and everything, in this context, not just "inventions". Think of it as creators or creations. I didn't write that, I quoted it. Natural persons excludes machines, monkeys, and everything else, except humans.

Yes, we can't copyright a single word or idea. But a set of words, like keywords are not the same as one word. You can't copyright a recipe, but you can copyright a book of recipes, as a collection. A set of words is not just a word, or no book could be protected, because it's just a bunch of words. My argument would be, that the description and keywords are ours and applied to a specific image. By scraping the images, they are just viewing them, but by using the associated words, that we created, to describe the image, there's something beyond fair use. (and yes I admit it's a stretch)

"each image is broken down to tokens and there's no way to reverse the process to identify where a token came from" Yes, and that's why we can't get credit for each new image. There is no one to one relationship from the original images to connect to the AI creations. Once trained the AI is on it's own.

I think Thaler intentionally went this way and more than once, has challenged, to cause a decision and bring Machine Learning into the court system as a test case. No one would try to register an art work as created by a machine, when they know the laws say, that can't be approved.

No I don't pay for the NY Times, sorry.  But yes, all of the points are going to be decided in the future and should be interesting.

Oh good, I found a free version:  https://artdaily.cc/news/161538/As-fight-over-AI-artwork-unfolds--judge-rejects-copyright-claim

thaler's rejected case was provocative =- claiming the ai software was the creator. 

Someone had to start the ball rolling. With no case law, there wouldn't be any review or appeals and it would probably just go down as, USCO says so. Because that's the way they interpret the laws. There needs to be a challenge to create change.