Mind-Blowing AI Image Generator - Give Visual Representation to C's Concepts?

I've started to use some AI tools for work to a) organise myself, remember stuff and b) edit writing

Is anyone making use of some of these new so-called AI tools regularly?

Adobe Photoshop beta (AI)

I have started to experiment a little bit with the new AI in Adobe Photoshop Beta... which is a highly mixed experience. First "Ah and Oh" - but then when you compare it with other tools - you realize how absolutely clumsy it is for creating new fantasy creatures.

When it comes to the creation of people, animals, and fantasy creatures, or let alone "planets" and "comets" - it is absolutely awful and most often pretty useless - especially compared with what @Elisasheva has shown off with Midjourney (the latter seem to give extremely fine detailed and well executed results for fantasy figures). Photoshop cannot touch this, not even to 1%.

What it can do
But there are situations, where Photoshop AI Beta works well, like with mice and rats, or changing hair in a portrait. Also, the way Photoshop beta AI, blends existing background with new elements in very realistic ways, such as light and shadows, are very well executed ! Also when eliminating certain areas in a background or photo works very well. (Which helps to correct things locally, for example when restoring old damaged photos) Even extending a photo beyond its borders, works interestingly well.

Mostly just for fun, such as comics
After a week, I manly use it for my "Helga comics" - which are like super silly, totally over the top hysterical and crude parodies, in which i change people's looks, sometimes into transmen/women-style caricatures. I've done this occasionally since the mid 90s, giving friends a good laugh. The new Photoshop AI beta, makes it a very effective tool in this regard.

At the same time - I noticed that i quickly get exhausted by the often far too clumsy AI results - and often wonder - what the F am i putting my energy into ? My primary usage seem to lean towards local corrections in images when needed. And for hysterical stupid cartoon creations. And that's about it.


Birthday Joke, 2023
"look who came over for a visit"

2023-01-17-22-39-32_3.jpg . 2023-01-17-22-39-32_4-Sal-birthday-with-rats-(AI).jpg


"I have a funny feeling about this..."

2023-01-17-22-36-44.jpg

2023-01-17-22-36-44-Ruth,-sitting-cat-with-Rat-3-(AI).jpg


The Swedish 'Addition'

My husband was complaining about his new fat extension around his hips (I call it the "Swedish addition", because life here is less hectic compared to life in Sicily, - so he has gained weight [ooops, now i said it]. In order to joke about with an AI image, i added a Tiger in Photoshop AI beta...

However, the 'Tiger' - meh - clearly wasn't interested, and just went away :-P

2023-05-27-11-57-27-Sal-(AI)-03.jpg

2023-05-27-11-57-27-Sal-(AI)-04.jpg


Hair & Animals

2019-10-25-22-05-48b_final.jpg

2019-10-25-22-05-48b_final-AI-08-Ruth.jpg

2019-10-25-22-05-48b_final-AI-13.jpg

Claus goes "Catherine Deneuve" (2004)

Through time I've liked from time to time, to make really absurdly goofy images in front of the camera, also with good friends, like you can see here my best friend Claus. The goofier, more hysterical, the better (and often i add talk bubbles, too, with cheap stupid talk)

So, i took that image recently, and changed parts of it with Photoshop AI in order to see what comes out of it. Many times, it gives strange unreal suggestions. But for fun comics, this isn't really a problem. Notice that the AI is pretty good in creating natural shadows, and blending AI elements naturally with the existing background. So far - the Photoshop AI beta - is more like a locally adapting, correctional image tool, rather than to create creatures from scratch, which it is incredibly bad at. When for example writing "comet in the sky" - you get the most stupid, totally unrealistic results in images. It's overall a pretty stupid AI, if you ask me.

But very funny, when you wish to make comic changes.

2004-11-28-17-11-16-Claus.jpg
original

2004-11-28-17-11-16-Claus-AI-02.jpg
 
Last edited:
"I have a bad feeling about this"

I just stumbled across this twitter entry, showing in an 1 minute video the way AI fuels, in increasingly convincing ways, many aspects of image and video. Showing how changes are done: such in movement, pose, expressions and layout of the objects in images. (I am sure that this also can be done for video) and i see a lot of problems with this, given how we consume and study the contents of media today... trying to figure out what pointings towards more honest reality - or just more lies and fakes, looking just too real.

At there same time, i also see just another example of how creators in offices live in a bubble (?) creating content that doesn't really exist, because with help of AI everything can be simulated and look "chic and convincing". More and more jobs are getting destroyed/replaced with this kind of virtual stuff, instead of using real people in the creation of art, ads, movies, documentations, well anything that has to do with visual contents... and keep in mind, all the texts written with help of, or entirely by AI... uh !

 
Adobe Photoshop beta (AI)

I have started to experiment a little bit with the new AI in Adobe Photoshop Beta... which is a highly mixed experience. First "Ah and Oh" - but then when you compare it with other tools - you realize how absolutely clumsy it is for creating new fantasy creatures.
What it can do
But there are situations, where Photoshop AI Beta works well, like with mice and rats, or changing hair in a portrait. Also, the way Photoshop beta AI, blends existing background with new elements in very realistic ways, such as light and shadows, are very well executed ! Also when eliminating certain areas in a background or photo works very well. (Which helps to correct things locally, for example when restoring old damaged photos) Even extending a photo beyond its borders, works interestingly well.

"Generative Fill"

Here is an article from DPreview, which explains with examples from the new Photoshop AI beta version - how the new tools works - e.g. how to fill in areas in a photo called "Generative Fill", or to remove areas in a photo (which often works very well, indeed), and to extend an existing photo's borders - which as well, works incredibly well (and can be handy as a photographical tool used with reasonable manner) But adding / replacing animals and faces... those just look goofy and weird, in my opinion. Good for Comics and "fun", but nothing serious. (Which perhaps is a good thing).



Vesuvio Crater Example (removing people)
3 June 2006

I tested the "Generative Fill" myself on a stitched panorama photo which i took on 3 June 2006 at the top at the Vesuvio Volcano near Napoli, Italy. Note: one person is twice visible in the original panorama, because she moved between the frames - and therefore turned up twice when i stitched the individual images together.

But the panorama is just a showcase of how Photoshop AI beta handles the removal of image elements in a photo: replacing and filling out existing areas in a natural way so it blends into the existing imagery. Indeed it does that job very well, I must say.

Note: I didn't remove all people in one go - but one by one - because the AI image is limited to 1024 pixel in length. If you select a much larger image area, you get very low resolution in the details. A way to bypass this, by selecting smaller areas in the image, and use "generative fill" one by one.

2006-06-03-Namnlöst2.jpg

2006-06-03-Namnlöst2b-Vesuvio-Il-Grande-Cono.jpg

PS Trivia:
I often notice that when I raised my camera in order to photograph a scene, and people are coming "into the way", they either stop or duck. I usually tell them: "No please, walk in like normal". I kind of like the unexpected, and don't mind at all when people walking into a scene i happen to take images of.

Perhaps i am a bit goofy in that regard, but welcome people to "walk into the picture" :-P
 
Let's talk about AI equivalents in other areas:
Fake your voice by Apple ™

6 June 2023

It is a bit ironic with all the AI getting ramped up in front of the public lately... Apple will with the next iOS 17 introduce to let you create a voice that sounds like you. Here the article from MacRumors. Of course they highlight this with the positive argument of "The feature is aimed at those who are at risk of loosing their ability to speak"... :rolleyes:


iOS 17 Lets You Create a Voice That Sounds Like You​

Monday June 5, 2023 3:37 pm PDT by Juli Clover

With the first iOS 17 beta, Apple has introduced a new accessibility feature called Personal Voice. First highlighted earlier this year, Personal Voice is designed to allow you to use artificial intelligence to create a replica of your voice.


ios-17-personal-voice-1.jpg

The feature is aimed at those who are at risk of losing their ability to speak, with Personal Voice offering these individuals the chance to "create a voice that sounds like them" for communication purposes.

Personal Voice is available in the initial ‌iOS 17‌ beta, so developers can begin testing it right away. It can be found under Accessibility > Personal Voice. Creating a Personal Voice is process that takes around an hour. Recording requires a quiet place with little to no background noise, with Apple instructing users to speak naturally at a consistent volume while holding the iPhoneapproximately six inches from the face.


ios-17-personal-voice-2.jpg


If there is too much background noise in your location, Apple will warn you that you need to find a quieter place to record.

Personal Voice requires you to read a series of sentences aloud, after which your ‌iPhone‌ will generate and store your Personal Voice. The Personal Voice can then be used with the Live Speech feature, which allows users to type-to-speak in FaceTime, the Phone app, and other communication apps.

Personal Voice will be available to the general public when Apple releases the first public beta of ‌iOS 17‌. Apple has said that ‌iOS 17‌ will be available to public beta testers next month.
 
Apparently, despite the failure of the metaverse, there is a lot of buzz about Apple's new Augmented Reality headset. Of course, some people see it as something marvelous but as always, the devil is in the details. The trend of addiction, surveillance, and behaviour control under big tech will be amplified. This video goes into some of the aspects on how the possible effects of this product may enfold:
 
I’ve only just begun digging into the links in this article that was shared by colleague who happens to be a software engineering architect:


“Research in artificial intelligence is increasing at an exponential rate. It’s difficult for AI experts to keep up with everything new being published, and even harder for beginners to know where to start.

So, in this post, we’re sharing a curated list of resources we’ve relied on to get smarter about modern AI. We call it the “AI Canon” because these papers, blog posts, courses, and guides have had an outsized impact on the field over the past several years.

We start with a gentle introduction to transformer and latent diffusion models, which are fueling the current AI wave. Next, we go deep on technical learning resources; practical guides to building with large language models (LLMs); and analysis of the AI market. Finally, we include a reference list of landmark research results, starting with “Attention is All You Need” — the 2017 paper by Google that introduced the world to transformer models and ushered in the age of generative AI.”

Seems like a good overview on and can help provide some insight on “how the heck did we get here?” 🤔😵‍💫
 
So ohoho ... this is 1st result of the command to Midjoureny Bot to imagine - Dies Sanctissimi Corporis et Sanguinis Domini Iesu Christi View attachment 75789
and this is the 2nd I typed a command again as new command haha ... my son recently told me that if you "think" kindly when you giving comand to Bot to make image for you, images will be much prettier hahaha ... so well maybe now Bot just wanted to please me to give me an image that I "wanted" instead of truth that Bot gave 1st time? hmmm TeslaSpark_Dies_Sanctissimi_Corporis_et_Sanguinis_Domini_Iesu_C_cb2bd94a-1a7e-470c-ab09-21c2c9...png
 
Adobe Photoshop beta (AI)

I have started to experiment a little bit with the new AI in Adobe Photoshop Beta... which is a highly mixed experience. First "Ah and Oh" - but then when you compare it with other tools - you realize how absolutely clumsy it is for creating new fantasy creatures.

When it comes to the creation of people, animals, and fantasy creatures, or let alone "planets" and "comets" - it is absolutely awful and most often pretty useless - especially compared with what @Elisasheva has shown off with Midjourney (the latter seem to give extremely fine detailed and well executed results for fantasy figures). Photoshop cannot touch this, not even to 1%.

What it can do
But there are situations, where Photoshop AI Beta works well, like with mice and rats, or changing hair in a portrait. Also, the way Photoshop beta AI, blends existing background with new elements in very realistic ways, such as light and shadows, are very well executed ! Also when eliminating certain areas in a background or photo works very well. (Which helps to correct things locally, for example when restoring old damaged photos) Even extending a photo beyond its borders, works interestingly well.

Mostly just for fun, such as comics
After a week, I manly use it for my "Helga comics" - which are like super silly, totally over the top hysterical and crude parodies, in which i change people's looks, sometimes into transmen/women-style caricatures. I've done this occasionally since the mid 90s, giving friends a good laugh. The new Photoshop AI beta, makes it a very effective tool in this regard.

At the same time - I noticed that i quickly get exhausted by the often far too clumsy AI results - and often wonder - what the F am i putting my energy into ? My primary usage seem to lean towards local corrections in images when needed. And for hysterical stupid cartoon creations. And that's about it.


Birthday Joke, 2023
"look who came over for a visit"

View attachment 75417 . View attachment 75416


"I have a funny feeling about this..."

View attachment 75418

View attachment 75420


The Swedish 'Addition'

My husband was complaining about his new fat extension around his hips (I call it the "Swedish addition", because life here is less hectic compared to life in Sicily, - so he has gained weight [ooops, now i said it]. In order to joke about with an AI image, i added a Tiger in Photoshop AI beta...

However, the 'Tiger' - meh - clearly wasn't interested, and just went away :-P

View attachment 75423

View attachment 75425


Hair & Animals

View attachment 75426

View attachment 75427

View attachment 75428

Claus goes "Catherine Deneuve" (2004)

Through time I've liked from time to time, to make really absurdly goofy images in front of the camera, also with good friends, like you can see here my best friend Claus. The goofier, more hysterical, the better (and often i add talk bubbles, too, with cheap stupid talk)

So, i took that image recently, and changed parts of it with Photoshop AI in order to see what comes out of it. Many times, it gives strange unreal suggestions. But for fun comics, this isn't really a problem. Notice that the AI is pretty good in creating natural shadows, and blending AI elements naturally with the existing background. So far - the Photoshop AI beta - is more like a locally adapting, correctional image tool, rather than to create creatures from scratch, which it is incredibly bad at. When for example writing "comet in the sky" - you get the most stupid, totally unrealistic results in images. It's overall a pretty stupid AI, if you ask me.

But very funny, when you wish to make comic changes.

View attachment 75429
original

View attachment 75430
Is it free? Can you provide a link please?
 
I was playing a little with Midjourney and decided to create some depictions of a secret meeting between rich and powerful and lizard beings. That's what I got.
I see a very dark story developing here - a story that begins with the main halls and rooms awash in gold, where rich people feast together with lizards, but the deeper you go down into the lower halls and basements, the weirder and scarier scenes, rituals, arrangements start taking place. And therefore lizards are starting to look more and more disturbing and terrifying. Interesting.

View attachment 75354

View attachment 75356

View attachment 75357

View attachment 75358
View attachment 75359
View attachment 75360
View attachment 75361
View attachment 75362
View attachment 75363
View attachment 75364
You have wonderful and terrible imagination :)
 
and this is the 2nd I typed a command again as new command haha ... my son recently told me that if you "think" kindly when you giving comand to Bot to make image for you, images will be much prettier hahaha ... so well maybe now Bot just wanted to please me to give me an image that I "wanted" instead of truth that Bot gave 1st time? hmmm View attachment 75790
Interesting how it draws that "church" style when recognize religion connectiion. As in "Dies Sanctissimi Corporis et Sanguinis Domini Iesu Christi".

Maybe it would be interesting if it can do it but in another style.
 
Interesting how it draws that "church" style when recognize religion connectiion. As in "Dies Sanctissimi Corporis et Sanguinis Domini Iesu Christi".

Maybe it would be interesting if it can do it but in another style.

I usually use that tools to "talk" with AI, and I wanted to see how AI sees "Corpus Cristi". And yes, interestingly, 1st I just copy pšaste Latin with no thoughts in my mind and doing some other stuff at the same time, and 2nd one I thought of Jesus and was focused on the result ... so it is REACTING with the frequency of my thoughts and it is indeed what is happening here with all AI tools, if we consider that all is energy frequency vibration, this what we now create or project with AI is the tuning in of what Tesla gave conceptual input for the Mind Projector explanation. So probably more STO-oriented players with AI can bring STO frequency to the visual surface? And in that respect giving inputs in that frequency of serving to others can create mood images of the same inspirational kind ... just my 5 cents ;)
 

Attachments

  • mindprojectortesla.jpg
    mindprojectortesla.jpg
    1.3 MB · Views: 4
"I have a bad feeling about this"

I just stumbled across this twitter entry, showing in an 1 minute video the way AI fuels, in increasingly convincing ways, many aspects of image and video. Showing how changes are done: such in movement, pose, expressions and layout of the objects in images. (I am sure that this also can be done for video) and i see a lot of problems with this, given how we consume and study the contents of media today... trying to figure out what pointings towards more honest reality - or just more lies and fakes, looking just too real.

At there same time, i also see just another example of how creators in offices live in a bubble (?) creating content that doesn't really exist, because with help of AI everything can be simulated and look "chic and convincing". More and more jobs are getting destroyed/replaced with this kind of virtual stuff, instead of using real people in the creation of art, ads, movies, documentations, well anything that has to do with visual contents... and keep in mind, all the texts written with help of, or entirely by AI... uh !

Wow, this tool is actually mind-blowing 😮
 
I tried Stable Diffusion. Much easier to use than Midjourney, no registering and "give us your data" stuff. Needs a bit of tunning to get what you want. Still strugles with people, but its OKish. First, I tried it to be symbolic and simple, but it was too dark, didnt felt good (those are in black and white), then I tried to make it more positive. And get carried away :)

Now me too thinks that there really is some kind of interaction. You must be in the "flow" to make pictures as you want them to look. Otherwise, they look generic and too bland. download.jpegdownload (2).jpegdownload (3).jpegdownload (4).jpegdownload (16).jpeg


download (17).jpegdownload (23).jpegdownload (25).jpegdownload (26).jpegdownload (27).jpegdownload (28).jpegdownload (29).jpegdownload (30).jpegdownload (32).jpegdownload (33).jpegdownload (34).jpegdownload (19)~2.jpegdownload (18)~2.jpegdownload (14)~2.jpeg
 
Back
Top Bottom