Midjourney: Difference between revisions

no edit summary
(Created page with "Midjourney is an artificial intelligence program capable of generating artistic images based on input text (i.e. prompts) <ref name="”1”">Claburn, T. (2022). David Holz, founder of AI art generator Midjourney, on the future of imaging. Retrieved from https://www.theregister.com/2022/08/01/david_holz_midjourney/</ref> <ref name="”2”">Borji, A. (2022). How good are deep models in understanding the generated images? arXiv:2208.107...")
 
No edit summary
 
Line 1: Line 1:
==[[Midjourney Photographers Styles Reference Sheet]]==
==[[Midjourney Artists Styles Reference Sheet]]==
==Overview==
[[Midjourney]] is an [[artificial intelligence]] program capable of [[AI image generation|generating artistic images]] based on input text (i.e. [[prompts]]) <ref name="”1”">Claburn, T. (2022). David Holz, founder of AI art generator Midjourney, on the future of imaging. Retrieved from https://www.theregister.com/2022/08/01/david_holz_midjourney/</ref> <ref name="”2”">Borji, A. (2022). How good are deep models in understanding the generated images? arXiv:2208.10760</ref>. It is also the name of the overall project; according to their official website, “Midjourney is an independent research lab exploring new mediums of thought and expanding the imaginative powers of the human species” <ref name="”3”">Midjourney Documentation. Quick Start Guide. Retrieved from https://midjourney.gitbook.io/docs</ref>. The AI-generating software has been compared to [[OpenAI]]’s [[DALL-E]] and likened, in its present form, to a social network for creating art based on text descriptions <ref name="”1”" /> <ref name="”4”">Florian, M-C. (2022). Sistemas de Inteligência Artificial como DALL-E ou Midjourney podem realizar tarefas criativas? Retrieved from https://www.archdaily.com.br/br/987358/sistemas-de-inteligencia-artificial-como-dall-e-ou-midjourney-podem-realizar-tarefas-criativas</ref>.
[[Midjourney]] is an [[artificial intelligence]] program capable of [[AI image generation|generating artistic images]] based on input text (i.e. [[prompts]]) <ref name="”1”">Claburn, T. (2022). David Holz, founder of AI art generator Midjourney, on the future of imaging. Retrieved from https://www.theregister.com/2022/08/01/david_holz_midjourney/</ref> <ref name="”2”">Borji, A. (2022). How good are deep models in understanding the generated images? arXiv:2208.10760</ref>. It is also the name of the overall project; according to their official website, “Midjourney is an independent research lab exploring new mediums of thought and expanding the imaginative powers of the human species” <ref name="”3”">Midjourney Documentation. Quick Start Guide. Retrieved from https://midjourney.gitbook.io/docs</ref>. The AI-generating software has been compared to [[OpenAI]]’s [[DALL-E]] and likened, in its present form, to a social network for creating art based on text descriptions <ref name="”1”" /> <ref name="”4”">Florian, M-C. (2022). Sistemas de Inteligência Artificial como DALL-E ou Midjourney podem realizar tarefas criativas? Retrieved from https://www.archdaily.com.br/br/987358/sistemas-de-inteligencia-artificial-como-dall-e-ou-midjourney-podem-realizar-tarefas-criativas</ref>.


Line 11: Line 16:
The program output, like DALL-E, results from AI models training on a large number of images. The third version of Midjourney has been introduced with enhancements in image quality coming from it’s social aspect, incorporating “a feedback loop based on user activity and response.” According to [[David Holz]], founder of Midjourney, “It's mind-bogglingly better and we didn't actually put any more art into it. We just took the data about what images the users liked, and how they were using it. And that actually made it better" <ref name="”1”" />.
The program output, like DALL-E, results from AI models training on a large number of images. The third version of Midjourney has been introduced with enhancements in image quality coming from it’s social aspect, incorporating “a feedback loop based on user activity and response.” According to [[David Holz]], founder of Midjourney, “It's mind-bogglingly better and we didn't actually put any more art into it. We just took the data about what images the users liked, and how they were using it. And that actually made it better" <ref name="”1”" />.


==Current applications and future==
==Current Applications and Future==
 
Midjourney has been adopted by both professional artists and amateurs. Indeed, the majority of people are using it for fun, engaging with their imagination. Some use it as a kind of art therapy, creating images of pets that have died, for example. Regarding professionals, architects and graphic artists use the program to help with the development of their concepts. It’s an easy and quick way to generate several variations of an idea to see what fits bests to the clients’ needs <ref name="”1”" /> <ref name="”3”" />.
Midjourney has been adopted by both professional artists and amateurs. Indeed, the majority of people are using it for fun, engaging with their imagination. Some use it as a kind of art therapy, creating images of pets that have died, for example. Regarding professionals, architects and graphic artists use the program to help with the development of their concepts. It’s an easy and quick way to generate several variations of an idea to see what fits bests to the clients’ needs <ref name="”1”" /> <ref name="”3”" />.


Line 24: Line 28:


==Development==
==Development==
Midjourney was founded by David Holz after leaving [[Leap Motion]], a company he ran until 2021. In 2022, an invitation only private beta was launched around early March <ref name="”1”" /> <ref name="”9”">Designboom. Midjourney. Retrieved from https://www.designboom.com/tag/midjourney/</ref> <ref name="”10”">Harmeet, G. (2022). Midjourney - Start your journey. Retrieved from https://weirdwonderfulai.art/general/midjourney-start-your-journey/</ref>. The open beta was announced on July 13th, 2022 <ref name="”11”">Twitter. Midjourney. Retrieved from https://twitter.com/midjourney</ref>.
Midjourney was founded by David Holz after leaving [[Leap Motion]], a company he ran until 2021. In 2022, an invitation only private beta was launched around early March <ref name="”1”" /> <ref name="”9”">Designboom. Midjourney. Retrieved from https://www.designboom.com/tag/midjourney/</ref> <ref name="”10”">Harmeet, G. (2022). Midjourney - Start your journey. Retrieved from https://weirdwonderfulai.art/general/midjourney-start-your-journey/</ref>. The open beta was announced on July 13th, 2022 <ref name="”11”">Twitter. Midjourney. Retrieved from https://twitter.com/midjourney</ref>.


==Creating images with Midjourney==
==Creating images with Midjourney==
After joining Midjourney’s Discord server, the user will be able to generate art based on text prompts. At first, the program will present four diffused images as previews. After the user chooses the ones he wants, a higher resolution version will be created. There is a limit of 25 free uses of text prompts, using the /imagine command and other queries, without subscription <ref name="”3”" /> <ref name="”10”" />.
After joining Midjourney’s Discord server, the user will be able to generate art based on text prompts. At first, the program will present four diffused images as previews. After the user chooses the ones he wants, a higher resolution version will be created. There is a limit of 25 free uses of text prompts, using the /imagine command and other queries, without subscription <ref name="”3”" /> <ref name="”10”" />.


Line 51: Line 53:


==Subscription and private mode==
==Subscription and private mode==
Without subscription, the prompts submitted by the users will be visible to all other that are part of the server <ref name="”10”" />. There are two subscription tiers, $10/month and $30/month which allow the user to submit text to the bot as a private Direct Message, receiving images as responses. The first tier gives access to 200 prompts and the second, unlimited access <ref name="”1”" /> <ref name="”13”">Bastian, M. (2022). Midjourney: DALL-E competitor enters open beta https://the-decoder.com/midjourney-dall-e-competitor-enters-open-beta/</ref>.
Without subscription, the prompts submitted by the users will be visible to all other that are part of the server <ref name="”10”" />. There are two subscription tiers, $10/month and $30/month which allow the user to submit text to the bot as a private Direct Message, receiving images as responses. The first tier gives access to 200 prompts and the second, unlimited access <ref name="”1”" /> <ref name="”13”">Bastian, M. (2022). Midjourney: DALL-E competitor enters open beta https://the-decoder.com/midjourney-dall-e-competitor-enters-open-beta/</ref>.


Line 57: Line 58:


==References==
==References==
<references />
<references />
370

edits