Rainman_666 said: AI generated images from various prompts asking for winter clothing, oceans, newly bathed, fighting with pastry, etc...
I've struggled getting the AI to get puffy clothing wet as easily as rainwear. I don't think it's a censor, I think it's because it doesn't have the source material from RW images so readily. The holy grail is puffy jackets and vests splattered with cream and cake batter.
do you mean like puffy dresses? if so try oversized ballgown it will puff up the arms and skirt im sure it might work with other things or it may just make them bigger
Rainman_666 said: AI generated images from various prompts asking for winter clothing, oceans, newly bathed, fighting with pastry, etc...
I've struggled getting the AI to get puffy clothing wet as easily as rainwear. I don't think it's a censor, I think it's because it doesn't have the source material from RW images so readily. The holy grail is puffy jackets and vests splattered with cream and cake batter.
Use a prompt that gives you this type of jacket and just add your prompts to it?
"showering happy tall woman with large breasts washing her hair, she stands deep in water, she is wearing a completely wet thick shiny short puffer down jacket over a completely wet black knitted turtle neck sweater, she wears soaking wet worn blue jeans. soapy water runs over her wet jeans that glisten in the light, and tight thigh high black leather boots, her boots are filled with soapy water and soap runs over her and water drips from her long wet hair. water showers down on her. highlights on her legs and hair. soap runs over her, and water drips from her long very wet hair. volumetric lighting, ultra photo realism, 35mm film quality, 8K, HD"
Occasionally I would get a warning for the prompt and occasionally for the image generated (which then wasn't shown.... )
Then I tamed it down (enough with the "large breasts"). Also, I tried various "curvy" and kept changing the _age_ That affected a lot. I never wrote "girl" (personally I think women in their 40s are the most attractive).
When I wanted to steer the thing into a particular scenario I would pick an image from a movie (or just whatever, even cartoons, just anything where I thought "hmm, if you added a lot of water and winter clothing... " ) with figures interacting in a way similar to what could work. Then I kept changing the fidelity to the source material.
Incidentally I noted that you don't really have to write out "attractive" or "gorgeous" as the internet probably is so saturated with "perfect looks" that the poor AI can't imagine anything but...
i do apologise if i am posting to many will have a break after this maybe
I stopped posting my image gens last week, given the volume I was creating, it felt I was cheapening the Forum. I'm not sure the forum is the place to dump a load of images, I started getting 100s of notifications and whilst the site is being reworked, I can imagine others are getting the same and the Wall being swamped. Hoepfully the site changes will make things a bit better all round
I felt the same quite quickly, so just post the ones I think are best from days worth. Lol
I've always wanted be a messy deliveroo cyclist, love the rainwear with a cycling helmet.
i do apologise if i am posting to many will have a break after this maybe
I stopped posting my image gens last week, given the volume I was creating, it felt I was cheapening the Forum. I'm not sure the forum is the place to dump a load of images, I started getting 100s of notifications and whilst the site is being reworked, I can imagine others are getting the same and the Wall being swamped. Hoepfully the site changes will make things a bit better all round
I felt the same quite quickly, so just post the ones I think are best from days worth. Lol
I've always wanted be a messy deliveroo cyclist, love the rainwear with a cycling helmet.
this sounds like a good idea maybe her bike broke and the terrain is deep sloppy mud impossible to cycle in. maybe shes at your door with your order in hand squashed with a messy smile
Punishment for late delivery, even by a few seconds, is to have cake batter poured all over you, second offence thrown into a slime pool. This is the gig economy, how else do we motivate the lazy bums to work nine hour shifts with no break?
messg said: I stopped posting my image gens last week, given the volume I was creating, it felt I was cheapening the Forum. I'm not sure the forum is the place to dump a load of images, I started getting 100s of notifications and whilst the site is being reworked, I can imagine others are getting the same and the Wall being swamped. Hoepfully the site changes will make things a bit better all round
I can certainly see where you're coming from, but I do think there's a danger of becoming overly concerned about this. The Wall is only one source of content on UMD of course and I imagine many people look at the various forums and groups directly as their main sources. I actually didn't even know the Wall even existed until pretty recently when MM mentioned it in a reply to me and, even though I do now, I make little use of it.
In terms of potential issues with the volume of AI content, I would make the following points:
1) There's no sign of significant criticism of the volume of AI content from either members who aren't into AI images or real-life WAM producers who might feel their material is being suppressed by disappearing off the Wall too quickly. In fact I haven't seen any criticism from producers at all, but I'm willing to be proved wrong!
2) The mods are having to approve all posts of AI images so, if they have concerns about the quantity of them, they would have said something by now to ask people to rein it in.
3) The biggest supporter of my image generations is a real-life WAM producer, so they don't seem to have any concerns, certainly at present.
As has been said, AI image generation is an exciting new area to explore and, like others, I'm eager to share what I'm generating in the hope that others will appreciate it too. I spend time, effort and money (on SeaArt) to do that and, if I had to significantly reduce what I posted, my interest and enthusiasm for sharing my stuff might start to wither and die.
I know that MM is working hard to implement filters that will give everyone control over whether or not they see AI images, at which point hopefully everyone who generates AI images can relax with regard to their productivity, but I would urge everyone not to get too concened about the volume of content in the meantime If things look like they're getting too out of hand, I'm sure the mods will let us know.
messg said: I stopped posting my image gens last week, given the volume I was creating, it felt I was cheapening the Forum. I'm not sure the forum is the place to dump a load of images, I started getting 100s of notifications and whilst the site is being reworked, I can imagine others are getting the same and the Wall being swamped. Hoepfully the site changes will make things a bit better all round
I can certainly see where you're coming from, but I do think there's a danger of becoming overly concerned about this. The Wall is only one source of content on UMD of course and I imagine many people look at the various forums and groups directly as their main sources. I actually didn't even know the Wall even existed until pretty recently when MM mentioned it in a reply to me and, even though I do now, I make little use of it.
In terms of potential issues with the volume of AI content, I would make the following points:
1) There's no sign of significant criticism of the volume of AI content from either members who aren't into AI images or real-life WAM producers who might feel their material is being suppressed by disappearing off the Wall too quickly. In fact I haven't seen any criticism from producers at all, but I'm willing to be proved wrong!
2) The mods are having to approve all posts of AI images so, if they have concerns about the quantity of them, they would have said something by now to ask people to rein it in.
3) The biggest supporter of my image generations is a real-life WAM producer, so they don't seem to have any concerns, certainly at present.
As has been said, AI image generation is an exciting new area to explore and, like others, I'm eager to share what I'm generating in the hope that others will appreciate it too. I spend time, effort and money (on SeaArt) to do that and, if I had to significantly reduce what I posted, my interest and enthusiasm for sharing my stuff might start to wither and die.
I know that MM is working hard to implement filters that will give everyone control over whether or not they see AI images, at which point hopefully everyone who generates AI images can relax with regard to their productivity, but I would urge everyone not to get too concened about the volume of content in the meantime If things look like they're getting too out of hand, I'm sure the mods will let us know.
They are fairpoints but I still think self moderation of volume in a forum setting is the best course of action. I'm a big advocate of AI and fascinated by how the usage has evolved over the last year both with ChatGPT, LLMs, Stable Diffusion and now DALLE and other image generators. I've been able to create some outstanding images despite the hurdles Dalle puts in front of us with content filters The fact remains though, it's bloody easy to generate masses of images. Using the API and other scripts, you can kick off a prompt and come back in 10 mins after a shower to a folder of a 100 images. And so.. this why I've held back posting more.
What the heck. I decided to give Bing a try. Several words were blocked, such as cleavage, revealing, low-cut. I had better luck with "custard" than with "pie" too.
It's frustrating. Its more fun watching the masters at work in this forum!
Yeah, Bing is very restrictive with their filters. And they seem to constantly change things, I have some prompts that previously worked for me but now get instantly blocked with a content warning.
WF1 said: Yeah, Bing is very restrictive with their filters. And they seem to constantly change things, I have some prompts that previously worked for me but now get instantly blocked with a content warning.
I kid you not when I say that I'll put something in and get a great image so I'll just press generate again and I get the dog thing saying it detected something.
BarryMcCockiner2 said: I'll put something in and get a great image so I'll just press generate again and I get the dog thing saying it detected something.
The opposite is true, too. I get the dog thing, click Create again and I get a good result.
Hey, this stuff is addictive. After the 6 I posted above, I stayed up and made another dozen.
Sleazoid44 said: The opposite is true, too. I get the dog thing, click Create again and I get a good result.
Yep, that's what I also do. Sometimes it's this way, sometimes the other way round.
But since a couple of days it's also true that the prompts from the archived pictures in my collection are not working anymore. Not even getting to the dog but being instantly warned not to write such nasty things if I do not want to get my account suspended. I'm not quite sure yet which part of the prompt triggers the filters, though.
if you get the dog on bing its because of the image generated not the prompt its self so it is possible to try a second time with the same prompt and you wont blocked. if something is wrong with what is written in the prompt it will block right away and you wont get a picture of a dog