|
Trying to make some test images to fool around with hueforge on my 3dprinter. Fingers crossed I can turn the top one into a bookmark but I'm not sure about my resolution I think I'll have to go with something much less detailed, but it would rip as an actual art print  
|
#
?
Jun 29, 2023 00:43
|
|
|
|
| # ? May 31, 2024 17:57 |
|
|
Tried to get ChatGPT to make ascii art >Draw Shrek in ascii art Certainly! Here's an ASCII representation of Shrek: ⠀⠀⠀⠀⢀⣀⡀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⣠⡖⠁⠙⢦⡀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⢀⡞⠁⠀⠀⠀⠈⠳⡄⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⢀⡾⠃⠀⠀⠀⠀⠀⠀⠈⢷⡀⠀⠀⠀ ⠀⠀⠀⠀⠘⢁⣴⠃⠀⠀⠀⠀⠀⠀⠈⠻⣄⠀⠀ ⠀⠀⠀⠀⠀⢳⠟⡀⠀⠀⠀⠀⠀⠀⠀⠀⢹⡄⠀ ⠀⠀⠀⠀⠀⠈⢷⡀⠀⠀⠀⠀⠀⠀⠀⠀⠈⠷⡀ ⠀⠀⠀⠀⠀⠀⠀⠈⠳⡄⠀⠀⠀⠀⠀⠀⠀⠀⠳ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⠁⠒⠒⠂⠈⠻⡆⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⡼⠁⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠁⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ >Sorry it is terrible Apologies for the quality of the previous ASCII representation. Here's an improved version of Shrek: ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣀⣠⡤⠤⠤⠤⢤⣀⣀⣀⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⡴⠛⠁⠙⠻⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⢷⡄⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⡾⠁⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠈⢿⡄⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣷⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠀⢀⣀⣀⣠⣄⣀⣀⡀⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⢰⣿⣿⣿⣿⣿⣿⣿⣿⣦⣤⣤⡀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠸⣿⣿⣿⡿⢿⣿⣿⣿⣿⣿⣿⣿⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠀⠈⠛⠁⠀⠈⠛⢿⣿⣿⣿⡿⠟⠁⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠀⠀⠀⠀⠀⠀⠀⠈⠉⠁⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣿⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⢿⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⢸⡇⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⢸⣧⣀⣀⣀⣀⣀⣀⣀⣀⣀⣀⣀⣀⣀⡀⠀⣸⡇⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠈⠛⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠁⠘⠁⠀⠀⠀⠀⠀⠀⠀⠀ >That looks nothing like Shrek . Just draw Shrek's face in ascii. Apologies for the previous attempts. Here's a simplified ASCII representation of Shrek's face ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣠⣶⣿⣿⣿⣷⣦⡀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢀⣴⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⣄⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣠⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡆⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⢠⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣧⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣦⠀⠀⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⠀⢰⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣆⠀⠀⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⢀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡄⠀⠀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣷⡀⠀⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⢠⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣧⠀⠀⠀ ⠀⠀⠀⠀⠀⠀⣠⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⠀⠀⢰⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⠀⢀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⢰⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⣾⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⠀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⢀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⣼⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⠀⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀ ⠀⢰⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀⠀ ⠀⣾⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⣿⡇⠀⠀⠀⠀⠀ ⠰⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠿⠃⠀⠀⠀⠀⠀ 
|
|
#
?
Jun 29, 2023 08:38
|
|
|
I want to see what a fork made of chainmail would look like and went down a different hole. chainmail fork on a plate, ominous lighting, foreboding texture  chainmail fork on a plate, ominous, foreboding lighting   chainmail fork on a plate 
|
|
#
?
Jun 29, 2023 13:42
|
|
|
This is a welded chainmail fork from the new trendy restaurant in town. MEEAT probably with an umlaut or something serves a smoked meat only menu in an upscale dining 'experience'. In a cave. You sort of palm the fork and use it like little pulled pork claws. Booked for months, make your reservations in advance. It's also expensive and not really worth it. But unless you want to be asked "Oh you haven't been?!" for the next year by people ordering Negroni Long Island Iced Tea's, you absolutely have to go just so you can tell people it was fine.
|
|
#
?
Jun 29, 2023 13:55
|
|
|
Used the last post as a prompt and got some tasty vittles from the restaurant. I got the Bone Broth and Braised Fat Bourguignon soup to start with some Angel Hair Pork.  And the Double Smoked Meatloaf Brisket Sandwich Au Jus for a main.  They were ok.
|
|
#
?
Jun 29, 2023 14:57
|
|
|
Sedgr posted:Used the last post as a prompt and got some tasty vittles from the restaurant. SLOP EM UP
|
|
#
?
Jun 29, 2023 16:13
|
|
|
really fancy cocktail with garnish and rim featuring beef broth e: okay this one is maybe drinkable looking really fancy cocktail with garnish and rim featuring beef broth, aquatic lighting, seaweed shading 
pixaal fucked around with this message at 16:25 on Jun 29, 2023 |
|
#
?
Jun 29, 2023 16:19
|
|
|
Jesus the first one is like a Bloody Mary/beef broth combo and it's disgusting. I love it.
|
|
#
?
Jun 29, 2023 16:26
|
|
|
I don't know why my meat cocktails made me laugh so much. Maybe because it was such a simple prompt. This is my favorite.
|
|
#
?
Jun 29, 2023 16:59
|
|
|
For today's meaty theme, I made gourmet sloppy Joe's dripping with sauces and juices photographed for a 1970s cookbook. Honestly, these all look fairly edible:   
|
|
#
?
Jun 29, 2023 17:20
|
|
|
|
|
#
?
Jun 29, 2023 17:31
|
|
|
I'd like it if someone could do a Renaissance banquet table still life but the food being served is a roasted xenomorph.
|
|
#
?
Jun 29, 2023 17:39
|
|
|
Dick Trauma posted:I'd like it if someone could do a Renaissance banquet table still life but the food being served is a roasted xenomorph. Not quite what you were hoping for, but still pretty  :   
|
|
#
?
Jun 29, 2023 17:45
|
|
|
|
|
#
?
Jun 29, 2023 18:13
|
|
|
went on a multi-model journey through the eyes of a confused cow going through a portal that is closing, beef rain, meat lighting      e: hi-res 
pixaal fucked around with this message at 19:58 on Jun 29, 2023 |
|
#
?
Jun 29, 2023 19:03
|
|
|
axolotl farmer posted:Tried to get ChatGPT to make ascii art Can you show me an example of complex ASCII art ChatGPT Certainly! Here's an example of a more complex ASCII art representation of a dragon: code:User Can you show me an example of Shrek complex ASCII art? ChatGPT Certainly! Here's an example of a complex ASCII art representation of Shrek: code: Can you describe to me what Shrek looks like ChatGPT Shrek is a fictional character from a series of animated films. He is a large, green ogre with a distinctive appearance. Here is a description of Shrek's features: Body: Shrek has a robust and muscular build. He is typically depicted as being quite tall and broad. His body is covered in green skin, giving him a unique and recognizable appearance. Face: Shrek has ... using the description of Shrek's features, can you draw him using ASCII art at least 16 lines tall ChatGPT Certainly! Here's an attempt at creating an ASCII art representation of Shrek using the description provided: code:User using the description of Shrek's features, can you draw him using ASCII art at least 16 lines tall including head, body and limbs ChatGPT Certainly! Here's an attempt at creating an ASCII art representation of Shrek using the description provided, including the head, body, and limbs: code:you only showed me the head. can you show me a picture of shrek with arms and legs added ChatGPT Apologies for the oversight. Here's an updated ASCII art representation of Shrek with arms and legs added: code:ChatGPT Certainly! Here's an ASCII art representation of Shrek that includes the head, arms, and legs, spanning at least 40 lines: code:
|
|
#
?
Jun 29, 2023 20:39
|
|
|
chatGPT wasn't designed for ASCII art and I wouldn't expect it to work in the foreseeable future. It's a large language model, not a large letters arranged to make pictures model, so it is just going to return examples of existing ASCII art from its training.
|
|
#
?
Jun 29, 2023 20:50
|
|
|
Yeah definitely pushing it pretty hard I was pretty happy to get two different, roughly face shaped results This was the result for guy fieri: edit: not working and various permutations I did get this one of a dog howling at the moon tho edit: nevermind
|
|
#
?
Jun 29, 2023 20:57
|
|
|
 became  all prompts here txt2img missed the tiling settings in the copy so I put a screenshot of them https://imgur.com/a/LqYzeyz It really focused in on the mountains.
|
|
#
?
Jun 29, 2023 21:23
|
|
|
messing around with adobe firefly's generative fill calmly doing some laundry 
|
|
#
?
Jun 29, 2023 21:31
|
|
|
Soulhunter posted:messing around with adobe firefly's generative fill LOL
|
|
#
?
Jun 29, 2023 21:44
|
|
|
Soulhunter posted:messing around with adobe firefly's generative fill oh you
|
|
#
?
Jun 29, 2023 22:47
|
|
|
        
|
|
#
?
Jun 30, 2023 05:49
|
|
|
https://i.imgur.com/cUMIQtK.mp4 would Earwicker fucked around with this message at 17:25 on Jun 30, 2023 |
|
#
?
Jun 30, 2023 16:10
|
|
|
i don't know if i would drink any of that
|
|
#
?
Jun 30, 2023 21:01
|
|
|
If I was interested in training a custom image gen model using only my own art (i.e. not merging a small custom model with existing models, but purely trained on my own images), about how many images should I expect to need? I know dreambooth and stuff like that can be done with 5-20ish images but as far as I understand it, that's injecting a small custom model "patch" type thing into existing full-featured SD models, right? Is this all more complicated than I'm making it out to be? To be more specific all I need it to do is one thing, like character designs - I don't need to train a model that can generate landscapes or buildings or whatever too (ideally I'd train separate models for those). Also I'm aware that most of the "work" will be in tagging those images, rather than gathering the images in the first place. The goal is to be able to generate images in my own art style that no one can say are trained on stolen art, or whatever. And then possibly sell a license to use that ethical model. deep dish peat moss fucked around with this message at 21:22 on Jun 30, 2023 |
|
#
?
Jun 30, 2023 21:16
|
|
|
The Conquest posted:i don't know if i would drink any of that That's clearly a creative appetizer and it's 6 wings in there you have to dig out messily. They give you a wetnap.
|
|
#
?
Jun 30, 2023 21:21
|
|
|
deep dish peat moss posted:If I was interested in training a custom image gen model using only my own art (i.e. not merging a small custom model with existing models, but purely trained on my own images), about how many images should I expect to need? I know dreambooth and stuff like that can be done with 5-20ish images but as far as I understand it, that's injecting a small custom model "patch" type thing into existing full-featured SD models, right? Is this all more complicated than I'm making it out to be? a full model is hundreds of thousands of images the 5-10 you are seeing is for a Lora, which still needs a larger model to go with it, preferably the same one the Lora was trained along side, if the model is too different the Lora wont function how you want at all.
|
|
#
?
Jun 30, 2023 21:28
|
|
|
Is that hundreds of thousands of images for a "full" model as in it can generate any type of image? I have over 1,000 random character/creature designs/sketches/etc that I've made over the past several years, like these:  If all I wanted the model to do was generate more similarly-styled creatures, would it need hundreds of thousands of images like that? Or would 1,000ish be fine for such a specifically-purposed model (with proper tagging, etc)? I'm mostly interested because the public stigma against AI images is the belief that it's trained on "stolen" art, which I think is an incredibly naive and short-sighted argument for a few reasons, but if I'm going to use AI for my projects I'd prefer to just avoid the stigma altogether and say "It's all trained on my own art". But I feel like to do that I'd need it to be more than a LORA and it would need to replace the use of the standard SD model in the first place and it sounds like that's not even feasible. Also because I see that as the main commercial viability of 'traditional' artists going forward: selling your art to (or as) AI models. I guess for that all I'd need is a LORA? deep dish peat moss fucked around with this message at 21:55 on Jun 30, 2023 |
|
#
?
Jun 30, 2023 21:45
|
|
|
Whatever else we will get out of this conversation, I think it'd run easier if we start on the same page about what's driving most* of the public outrage still: fear of losing hypothetical money. A fear that a persons paid work would be done for free by a machine. You should fully expect that solving the valid concerns about theft, amorality, the classical ghost stories about dangerous dystopia machines etc, will not stop the outrage. You won't calm the folks who liked being the only one who could do the work and are trying to argue around that fact or hide that's whats making them mad. For all the hard to describe effects this tech will have on our futures, it is the money that kicked over the hornets nest *concerns about gatekeeping behind massive coroprations, or independent artists who are afraid people won't love their smol paintings anymore, is related but not the same. Everyone itt supports them for one
|
|
#
?
Jun 30, 2023 22:51
|
|
|
ThisIsJohnWayne posted:Prattle Sir, this is a Wendy's.
|
|
#
?
Jun 30, 2023 23:04
|
|
|
I've been busy making art and writing for two projects and managed to get a decent RunwayML Gen-2 AI video teaser trailer done for one of them this afternoon. It was supposed to be just a couple clips thrown together but I ended up making it like a whole TV show intro. Used royalty free music from White Bat Audio on YouTube, he does awesome John Carpenter style stuff for anyone to use with credit. https://i.imgur.com/EsRWpoK.mp4 YouTube: https://www.youtube.com/watch?v=2ZqJblB_rvU Roman fucked around with this message at 03:11 on Jul 3, 2023 |
|
#
?
Jun 30, 2023 23:36
|
|
|
quote is not edit
|
|
#
?
Jun 30, 2023 23:38
|
|
|
meat cocktail as a megadeth album cover with like conspiracy zombies and stuff
|
|
#
?
Jul 1, 2023 01:32
|
|
|
hydroceramics posted:For today's meaty theme, I made gourmet sloppy Joe's dripping with sauces and juices photographed for a 1970s cookbook. Honestly, these all look fairly edible: This is a style? Because it unlocks a core memory for me from early to not quite late 90s in Finland. Maybe cookbook my parents had, possibly one with cards with a picture for each recipe? Maybe grocery store with similar pictures strewn around. The lighting style is unmistakable.
|
|
#
?
Jul 1, 2023 03:34
|
|
|
At the end of all this AI stuff is the death of monoculture giants like Disney. AI is already out, invasive, cheap. I can, and do, right now, generate texts, stories, dialogs, on my own computer. Music, movies, pictures. It's just going to get easier to do this, cheaper, more powerful. It's already over, the shift has happened and we just haven't gone forward enough in time to realize it. Roman posted:specific characters https://github.com/s0md3v/roop https://github.com/s0md3v/sd-webui-roop roop, both a video deep-fake generator (and all the uncomfortable things that come with it) and a stable diffusion plugin allowing the same function on images. What this allows you to do is what people have been asking for a long time now: Generate a single character and then bring them back up. This native american CEO boss lady I used MidJourney to create, she isn't anyone in particular, but distinctive:  Using roop in stable diffusion she's now:  An 80s stylish business woman!  A race car driver!    Shopping!  Making cookies!    A man!  Younger!  A space marine or cosplayer, your choice!  LARPing!  A scientist!  Boxing!  At the club!   Scaring the children! All done as quick as any other image in Stable Diffusion, with roop running just afterwards to put the face on, it looks like this:  That single image imposed that face on all the images I just posted. One picture, seconds of time, a default character you can summon up.
|
|
#
?
Jul 1, 2023 04:02
|
|
|
KakerMix posted:At the end of all this AI stuff is the death of monoculture giants like Disney. AI is already out, invasive, cheap. I can, and do, right now, generate texts, stories, dialogs, on my own computer. Music, movies, pictures. It's just going to get easier to do this, cheaper, more powerful. lmao why would cheaper texts, dialogue, pictures, music and so on be the "death" of disney? its going to hurt a lot of artists who work for companies like disney, sure. the company itself, unlikely.
|
|
#
?
Jul 1, 2023 05:05
|
|
|
Earwicker posted:lmao Disney hurts artists plenty already. These tools keep getting cheaper and more powerful the old ways of doing things become just another way, not the only way. It's happened before with TV, radio, the internet, I can't see why it can't happen again with fully-AI generated content. We've seen trial runs of it already and people willing watch it, and we aren't even a year in yet. No Disney needed, people do this poo poo without a giant studio. I'm sure Disney will still be around crushing VFX artists even with AI tools.
|
|
#
?
Jul 1, 2023 06:04
|
|
|
KakerMix posted:it's happened before with TV, radio, the internet, I can't see why it can't happen again with fully-AI generated content. none of these technologies resulted in the death of powerful media conglomerates. quite the opposite. quote:No Disney needed, people do this poo poo without a giant studio. yeah and people can make albums without a recording studio but universal music group still exists and made 11 billion last year. and spotify made 12 billion. innovations like streaming help a few rich assholes get richer while most of the people actually making the music get less and less. i dont see ai helping that situation at all. dont get me wrong, ai can do some cool stuff and i enjoy loving around with the tools and making stuff, im not some luddite crusader, but i do think the idea that its going to turn the tables on a company like disney seems very naive. Earwicker fucked around with this message at 08:28 on Jul 1, 2023 |
|
#
?
Jul 1, 2023 08:18
|
|
|
|
| # ? May 31, 2024 17:57 |
|
|
I think your taking it too much at face value I'm reading it as Disney isn't the only one that can make use of it. Which yeah you can run this on your computer but you can run everything related to current animation too. At some point like how cgi eventually required more people over time to get a better product, we will see a Disney movie made with hundreds of people using AI tools. It's a small window where new companies can rise. CGI gave us DreamWorks. We are going into that window with AI right now
|
|
#
?
Jul 1, 2023 09:21
|
|