# Midjourney AI: Text To Image Supercharged! ## Метаданные - **Канал:** Two Minute Papers - **YouTube:** https://www.youtube.com/watch?v=jbe6t4GiljU - **Дата:** 17.07.2023 - **Длительность:** 5:28 - **Просмотры:** 82,343 ## Описание ❤️ Check out Weights & Biases and sign up for a free demo here: https://wandb.com/papers Sources: https://twitter.com/camawjones/status/1673428460981583872 https://aituts.com/midjourney-prompt-weights/ https://twitter.com/antman_Ai/status/1676168501004738560 Camels original prompt: Momsuca Cyberpunk original prompt: artefaktik My latest paper on simulations that look almost like reality is available for free here: https://rdcu.be/cWPfD Or this is the orig. Nature Physics link with clickable citations: https://www.nature.com/articles/s41567-022-01788-5 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Aleksandr Mashrabov, Alex Balfanz, Alex Haro, Andrew Melnychuk, Benji Rabhan, Bret Brizzee, Bryan Learn, B Shang, Christian Ahlin, Eric Martel, Geronimo Moralez, Gordon Child, Jace O'Brien, Jack Lukic, John Le, Kenneth Davis, Klaus Busse, Kyle Davis, Lorin Atzberger, Lukas Biewald, Martin, Matthew Valle, Michael Albrecht, Michael Tedder, Nikhil Velpanur, Owen Campbell-Moore, Owen Skarpness, Rajarshi Nigam, Ramsey Elbasheer, Richard Sundvall, Steef, Taras Bobrovytsky, Ted Johnson, Thomas Krcmar, Timothy Sum Hon Mun, Torsten Reil, Tybie Fitzhugh, Ueli Gallizzi. If you wish to appear here or pick up other perks, click here: https://www.patreon.com/TwoMinutePapers Thumbnail background design: Felícia Zsolnai-Fehér - http://felicia.hu Károly Zsolnai-Fehér's links: Twitter: https://twitter.com/twominutepapers Web: https://cg.tuwien.ac.at/~zsolnai/ #midjourney ## Содержание ### [0:00](https://www.youtube.com/watch?v=jbe6t4GiljU) Introduction Dear Fellow Scholars, this is Two Minute  Papers with Dr. Károly Zsolnai-Fehér. Today we are going to have a look  at the newest features Midjourney,   a text to image AI offers. Note that this  somewhat similar to Stable Diffusion,   an open text to image AI that we can run at  home for free, but here, a subscription is   required. Some of them cost quite a bit. So, is  it worth it? And what do we get for this money? ### [0:30](https://www.youtube.com/watch?v=jbe6t4GiljU&t=30s) Quality Well, I would say two words:  quality and convenience. Let’s talk about quality first. If we asked  Midjourney version 1, just a bit more than a   year ago for a scene with desert dunes, and  camels in front of The Great Sphinx of Giza,   we got something like this. Or this. But,  when asking version 5 to create the same,   we get this. Now that is an incredible improvement  in just one year. However, in the prompt, we also   asked for a highly-stylized result, and it seems  to have ignored that. Also, these images have   some artifacting, and the camels feel a little  copy-pasty. Hmm, yes. It’s almost as if just one   or two camel types were copied over many times.   No matter, because version 5. 2 is already out,   and it fixed that too. And the results  are breathtaking, stylized results. Loving it. And when asking for a high-contrast  surreal collage, with version 5,   we got something like this. While version 5. 2 is  a little more imaginative. I like this one a great   deal better. Now, as a counterexample, when asking  for vibrant california poppies, the results are a   toss up. These could easily pass for two images  from the same model. And both are excellent. So this was about quality, and Midjourney  has a great deal to offer when it comes to   convenience. For instance, whenever we create  an image, and we are happy with the result,   but not with the framing, we can now also ask it  to zoom out. If we have an eye, we can say okay,   but whose eye is this? And there we go. Fantastic.   And we can not only zoom, but we can also do some   panning. Little AI, tell me, what’s next to this  creepy guy? Oh yes, of course, another creepy guy. And there are a tons more  convenience features in this,   for instance, it supports prompt weights.   For instance, if we ask for a wooden teapot,   we can even put more emphasis on the wood  part. And we get this. More like wood,   less like a teapot. And then even more. Note  that the free Stable Diffusion can also do this. It also supports one of my favorites  for productivity: permutation prompts. ### [3:09](https://www.youtube.com/watch?v=jbe6t4GiljU&t=189s) Multiple Prompts This is really underrated. If we are sure  that we wish to get a painting of a bird,   but we are not quite sure which bird,  we can add multiple birds in the same   prompt. This will give us all four bird types  without having to create four prompts for it. And for the first sight, this does not seem like  a big thing, but this really is. Especially that   Midjourney supports multiple permutations, and  even nested ones. With this, we can try out   tens of ideas with just one prompt. An incredible  multiplier for your brain. I absolutely love this. And finally, it can also help us shorten our  prompts. What is that? Well, if we have tried ### [3:56](https://www.youtube.com/watch?v=jbe6t4GiljU&t=236s) New Prompts out way too many ideas and have a big bloated  prompt, it can help us sort out what parts of this   actually matter. Then, it gives us 5 new prompts  that are progressively shorter and shorter,   but all of them will try to retain the  heart and soul of the original one. So, these are some of the cooler features  of Midjourney, there are so many of them,   for instance, we haven’t even talked  about Remix mode, and more. So,   what do you think? Are you using  Midjourney? Is it worth the money,   or is it free Stable Diffusion all the way  for you? Let me know in the comments below! ### [4:39](https://www.youtube.com/watch?v=jbe6t4GiljU&t=279s) Outro Thanks for watching and for your generous  support, and I'll see you next time! --- *Источник: https://ekstraktznaniy.ru/video/13102*