In the news…a new service https://www.deviantart.com/team/jou...ntArt&utm_medium=Journal&utm_campaign=DreamUp More: https://www.cbr.com/deviantart-ai-rights-change-user-complaints/ https://stormtrack.org/community/threads/make-auto-generated-tornado-images-here.32281/#post-375274 This looks like watercolor https://www.artstation.com/artwork/aGqWvR like charcoal https://www.pinterest.com/pin/559853797438562147/ Wood? https://www.reddit.com/r/BeAmazed/comments/ygjk9y/a_carved_wooden_sculpture_of_a_submerged_man_with/ A whole film has been done in A.I. https://interestingengineering.com/culture/megastructure-orbital-ring-around-earth-film Now we have this to look forward to--T2CI GAN: Researchers at the Computer Vision and Biometrics Lab of IIT Allahabad and Vignan University in India have recently developed a new GAN-based model that can generate compressed images from text-based descriptions."The idea of T2CI GAN is aligned with the theme of 'direct processing/analytics of data in the compressed domain without full decompression,' on which we have been working on since 2012," Mohammed Javed, one of the researchers who carried out the study, told TechXplore. "However, the idea in T2CI GAN is a bit different, as here we wanted to produce/retrieve images in the compressed form given the text descriptions of the image." https://techxplore.com/news/2022-10-t2ci-gan-deep-compressed-images.html Further refinement: Yilun Du, a Ph.D. student in the Department of Electrical Engineering and Computer Science and affiliate of MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL), recently developed a new method that makes models like DALL-E 2 more creative and have better scene understanding. https://techxplore.com/news/2022-10-ai-image-robots.html DALL-E is now this https://www.craiyon.com/ https://www.engadget.com/deviant-art-is-launching-its-own-ai-art-generator-140006449.html a trend? https://www.nytimes.com/2022/12/31/opinion/sarah-andersen-how-algorithim-took-my-work.html https://nypost.com/2022/12/29/dystopian-new-tech-can-doctor-footage-for-any-smart-cam/amp/ https://www.nextbigfuture.com/2023/03/midjourney-image-generating-ai-improves.html
And now, this. https://arstechnica.com/information...odels-from-written-descriptions-thanks-to-ai/ sigh.
For a pretty measured look at the implications of what the current trend of AI-generated “art” means not only for artists but the human race in general, I found this essay by illustrator Steven Zapata pretty eye-opening. It's a bit long-ish, but I recommend every artist to listen to this. The TL;DR version is that the future is pretty bleak for artists and that we should actually try to fight it as good as we can. It's practically the end of creativity and wants to be a solution for a problem the human race never really had. Because why would we want computers to take creating art from us? Creating art is work, sure, but it's fun work. You shouldn't want a computer taking that away from you.
Watched the vid, thanks for the link, great vid! I’d like to discuss it, but feel like maybe a new thread would be appropriate as I feel like we’re hijacking the thread.
Perhaps a bit more on topic--here is a tool that turns sketches into code: https://techxplore.com/news/2022-11-tool-code.html Maybe that can be linked to give the art-bot's a nudge. They still don't like Star Trek ships and keep turning them into naval craft--so by inputing/tracing ship blueprints--it might give them a shove. Finding the right A.I. for you: https://techxplore.com/news/2022-12-ai.html On language https://medicalxpress.com/news/2023-01-voice-brains-sight.html Using wi-fi for human figures https://techxplore.com/news/2023-02-wi-fi-3d-human-meshes.html Landscape scans https://techxplore.com/news/2023-02-faster-accurate-3d-tool-recreates.html Frida the paint-bot https://techxplore.com/news/2023-02-ai-powered-frida-robot-collaborates-humans.html
If you haven't, watch the vid. I toyed with the idea of using one of these to help with things like backgrounds, after watching the vid, nope. Everything you type into a prompt is recorded by the AI. It's recording, assessing and learning from the data. The time is coming when, based on all the information it's gathered, seen what's trending, it will generate art on it's own without any need for a human being. And it will vomit forth an ocean of images, drowning out any hope you ever had of getting your work seen by enough people to matter. We're already seeing the beginnings of this with human input information. DeviantArt is bombarded by AI images. One of the easiest ways to tell if it's AI is by the lack of any commentary beneath the image. What inspired them, what tools they used, etc. Because when you can sit around in your underwear and just plunk words into a prompt because you're bored and want to see what the machine spits out, there's no real need to share any insight into your thinking or feelings on the matter because there were none. Apologies to the OP, I realize you wanted to share some images and your thread turned into something else.
A.I. may also be used in assembling models and other tech: https://techxplore.com/news/2022-12-automated-thousands.html But it can help us detect fakes too, at least https://techxplore.com/news/2022-12-ai-app-deepfake-videos-famous.html Now to get a 'bot to fix nacelle-droop... https://headtopics.com/us/ai-learns-to-write-computer-code-in-stunning-advance-32646647 Now this advance: https://www.theverge.com/2022/12/8/...ty-chatgpt-stable-diffusion-commercialization -might allow a running conversation if tied to the art-bots: “no—not a navel vessel…nacelles please…higher…” Watch hacking https://techxplore.com/news/2023-01-modification-shor-algorithm-powerful-quantum.html And now—Ghostwriter https://arstechnica.com/information...nted-ai-powered-typewriter-that-talks-to-you/ The coming year https://www.nextbigfuture.com/2022/12/will-ai-and-space-breakthroughs-continue-in-2023.html https://www.nytimes.com/2023/01/07/technology/generative-ai-chatgpt-investments.html https://phys.org/news/2023-01-artificial-intelligence-gain-insights-personality.html DFAI https://techxplore.com/news/2023-01-dfai-piece-artificial-intelligence.html Point-E https://techxplore.com/news/2022-12-openai-point-e-machine-quickly-3d.html
The latest advance concerns voices https://techxplore.com/news/2023-01-microsoft-vall-e-faithfully-voice.html A team of researchers at Microsoft has demonstrated a new AI system that is capable of mimicking a person's voice after training with a recording just three seconds long. The team explains developing the new app in a paper published on the arXiv preprint server. https://valle-demo.github.io/ On languages https://techxplore.com/news/2023-01-automatic-speech-recognition-languages.html PDF search? https://github.com/ocrmypdf/OCRmyPDF The latest https://techxplore.com/news/2023-01-dall-e-reveals-human-creativity.html
Awesome stuff. I am sure this technology humanity really needs is going to be used resbonsible and not to scam the elderly and confused, or spread misinformation to finally get rid of democracy. Perfect time to come up with this stuff. /s Isnt it fantastic to have grown up when TNG first aired...makes the current state of humanity so much more nightmareish.
I've been playing with one of these (albeit a far less advanced and more crusty version), they're pretty amazing. I'm using it to create custom voice lines for a Half-Life mod (which will be free) Of course, it also means we're heading into a time where we won't have a clue what's real and what isn't.
Ugh---Space News is using AI https://spacenews.com/buckle-up-it-...ys-vaunted-resilience-will-be-tested-in-2023/ Fie on them! More on the subject https://techxplore.com/news/2023-02-ai-based-image-generation-generate-trainer.html Now it's music: https://techxplore.com/news/2023-01-google-musiclm-hi-fidelity-music-generating.html At least you can search for medical images without being an expert now ttps://phys.org/news/2023-01-software-enables-automated-analysis-biomedical.html
edit: prompt: analog polaroid photo of one 29 year old woman in Star Trek Uniform, pale skin, wild hair, long hair, strong wind blowing hair, upper body, dark hair, detailed skin, detailed eyes, realistic eyes, 20 megapixel, detailed skin, detailed face out of maybe around 100 tries, these two came out pretty close of what i wanted edit2: bonuspic
civitai-model: RealisticVision.1.3 and civitai-model: deliberate so i'm done playing with different models of stable diffusion for now, it is very interesting. i know that both musicAI and animation/videoAI are coming full speed. we have the chatGPT for stories and with picture/graphic making AI's like different stable diffusion models... where will this lead?.... i am almost certain, within a year, i can push a handfull of buttons and "make" a new star trek movie.
Interesting. Were all the prompts for "Star Trek uniform" always presenting a uniform with the arrowhead emblem?
yes, the arrowhead insignias were always there in more or less the correct shape, i didn't have to specify that other than "star trek uniform" in any of the models.