Turn any idea into an image in seconds with Midjourney AI!
HTML-код
- Опубликовано: 22 дек 2022
- A full tutorial on how to create AI Avatars, app icons, and realistic images with Midjourney AI. Efficient prompting, and making your images more predictable, more realistic, or more cartoonish. How to use existing images as input for AI generation. Don't want to deal with setting up Stable Diffusion, and don't want to be limited by what Lensa AI or Jasper Art are offering you? Use Midjourney!
Join our Discord to try out Midjourney for yourself for free.
/ discord
More tools:
GPFGAN - to fix faces: bit.ly/3VlzIbd
img2prompt - reverse an image: bit.ly/3veUkap
More tips for prompting:
Olivio Sarikas Channel (A bunch of prompting tips): bit.ly/3VkIjek
PiXimperfect - Last selfie on Earth: bit.ly/3GgcM8R
More material (How this works)
Videos:
Vox - AI Art: bit.ly/3FUqooP
AssemblyAI - Diffusion Models explained in 4 levels: bit.ly/3C12Ldb
Articles and papers:
Or Patashnik's papers: orpatashnik.github.io/
AssemblyAI's "Introduction to Diffusion Models": www.assemblyai.com/blog/diffu...
Annotated Diffusion Model: huggingface.co/blog/annotated... 
Наука
Keep it up with videos like this one and in no time you'll be the kind of midjourney tutorials.
How lucky we are to witness these technological breakthroughs and be amongst the first explorers!
Absolutely! These are exciting times! But I prefer to cover all kinds of digital life hacks, not just Midjourney - we live in a marvelous time, and there's a lot of exciting technology around, and I can't wait to talk about all of it 😀
This was incredible my man! Thank you SO MUCH for the last 2 tools you shared! 🥰
Glad you enjoyed it! More digital life hacks are coming soon to this channel, working on a number of exciting videos.
This is the type/quality of tutorials I have been waiting for. Bravo
Glad you liked it! More to come!
One of the best videos I’ve seen for helpful tips for prompts. Thank you for such informative content!
It was also interesting to hear the background how algorithm work
Glad it was helpful!
This is it! The cleanest and solid tutorial for Midjourney, thank you
Glad it was helpful!
That was an excellent introduction to how it all works and how to do some fantastic pictures in Midjourney. Thank you!
Glad you enjoyed it!
Best Midjourney video I have seen to date. No one else is giving us these amazing tips. Will be back to reference this video 1000 times from now.
Thank you so much! I really appreciate it!
Thank you! I learned a lot please do more tutorials and go into depth on the discussion at the end of this one!
I'm glad you learned! Not sure what discussion you mean - you want me to share more addon tools?
this is amazing. I didn't know Midjourney was capable of all this. Thank you!
Just wait to see what else Midjourney is capable of, and Stable Diffusion is capable of even more!
The best explanation all over RUclips .... you ARE THE BEST !!!
Wow, thanks! I'm glad you liked it!
WOW!!!!! This was amazing!! Thank you so very much. This really helps!
Glad it was helpful!
Surprisingly informative, thanks mate.
Glad it was helpful!
this was very helpful and straightforward-Thank you!!
Glad it was helpful! You're very welcome!
love this tutorial. have watched many times, learned alot
Thank you so much, it warms my heart to read this!
This is very insightful, thank you so much for your tips!
You're so welcome!
I'm SMH in awe, while Liking and Subscribing. Amazing video and great explainer
Welcome aboard! Super glad it was helpful.
Thank you so much for this useful tutorial. I am actually listening to it as i apply your awesome tips while having midjourney open
Glad it was helpful! Would love to see the beautiful images you create! I'm always hanging out in the Discord in case you have any questions.
@@hml thank you for the invite. I am re-watching this particular video again lol
Thanks for the explanation of the science behind AI for images. Subbed.
Thanks for the sub! I'm glad the explanation was clear, and you enjoyed the video!
Great and comprehensive introduction to midjourney. I was able to follow along while using midjourney for the first time. Thanks David!
Great to hear! Didn't see you in our discord, following along, though :-)
@@hml Joined! Apparently, I did not follow along well enough 😂
I'll need a second watch through!
Very interesting tutorial! Thank you very much!
You are welcome!
Great tips, thank you!
You are welcome!
Spectacularly help video! Thank you so much, subscribed!!
Thank you! I'm glad it was helpful to you!
So far the most useful guide for these AI tools....a Biiiig Thank you🙏🙏
Glad it helped! Any other tools you'd like for me to make a guide for?
I am from Bangladesh. I have been searching for a video like this for many days. Its very clear that you know a lot. Hoping more videos like this. Btw Thank you.
Thanks and welcome!
This is amazing!!
The tech is literally science fiction!
I have subscribed your channel. I must say you are the best and deserve wider reach
Thank you so much 😀 It's comments like this that keep me motived, and inspired to create more!
Thanks a lot. Great, to the point content, very well presented!
Glad you enjoyed it! Another detailed Midjourney video is coming in a few days.
Awesome vid, thanks man 💪🏻
Thank you, man!
Great video, very well explained. Thanks very much
Glad you enjoyed it! Thank you!
This is such a life saver ! Especially for someone like me who is a content creator. Thank you for this information. I am currently trying to learn more about a new one called BlueWillow. Please do that one next !
I'm happy to say that Blue Willow is now available in the "Hacking Modern Life" Discord, and I will do a video on Blue Willow soon as well.
Very informative and condensed. Could you make more vidoes on Midjourney usages?
Hi! Thanks for your comment. I'm actually releasing another detailed video on Midjourney v5 today, so stay tuned :-)
this is pretty cool thanks for sharing this!!
Glad you liked it!
@@hml thanks again managed to create some photography related stuff thanks to your tips
Great tips about the other two sites also, big thanks
Glad it was helpful! I'll share more tools like that in the future!
GREAT JOB! SUPER INFORMATIVE!
Thank you so much, glad you enjoyed it!
Amazing no fluff video thanks ❤
You're absolutely welcome!
Very informative video! Thank you!
Glad it was helpful!
Wow ❤ so much valuable information. Thank you 👍😊
My pleasure 😊 I'm glad it was helpful!
posting this here because my Discord keeps freezing(Maybe updating my Windows OS will fix the issue, i am still using Window 8) :
HI David I hope that your current moment finds you well. Although i am taking a short break from Midjourney(refocussing on metallurgical art projects)I still intend to continue viewing your RUclips channel. Your advice, list of alternative AI generators ,as well as techniques utilizing other digital applications in conjunction with art generators are a gift. Thank you for sharing your personal time to instruct us on this ever evolving technology. Watching your videos is akin to viewing skillshare classes, except your classes are more genuine, easy to absorb, get to the point quicker and have a more engaged with the students watching.
Thank you so much! It's such a pleasure reading this, and I will definitely keep on sharing what I learn about modern life hacks, including generative AI, as I really enjoy navigating this new world. Hope to see you in my future videos!
Really good explanations, thx alot
Glad it was helpful!
What a great video!!
Thank you so much!! Working on more videos right now.
This video was an excellent explanation and example
Glad it was helpful! Thank you for watching!
Till this video, I was impressed with what is possible, but this is just mindblowing. When you started working with photos from yourself, I couldn't believe it. How can I ever trust a photo that I see when this gets to the point of photoquality perfection?
There are still a bunch of ways to detect AI generated images - pay attention to the hands, repeating patterns in the hair, uneven skin... But yeah, it gets crazy realistic. If you haven't already, check out my video on Midjourney v5, that have became even more realistic!
Great content! Thank you!
Glad you liked it!
Great job, very informative🙌🙌🙌
Glad you think so! Was trying to pack as much info as I could in the shortest amount of time.
thank you for the excellent video!
Glad you liked it!
man i like u for sharing such amazing things
It's my absolute pleasure! Hopefully you'll find many other useful things on my channel.
Outstanding information. MJ is pretty geeky. You did a great job explaining some of it. THANKS
Glad it was helpful! It is a challenge to balance "going too deep" and making it complicated, vs not making it clear at all, but I'm going to try to keep the balance.
fantastic video, thank you very much
You are welcome!
Very informative, Thank you.
Glad it was helpful!
Great tutorial :)
Glad you liked it!
Wow!... this is amazing... I need to become a master with this
Oh yeah, definitely! But if you want to have more control over the images you generate, look into Stable Diffusion, and I'm working on a whole series about it.
You are a master! thanks for sharing.
Wow, thanks! It's my pleasure.
What a great video! Thank you for sharing your knowledge🙏🙏🤗
You are so welcome!
@@hml Many Thanks for your hardworks! 💪💪Appreciate very much 🙏🙏😊😊
Fantastic !!! 🎉 Thank you
Thank you! Cheers!
one of the best explained video
Thank you, I'm glad it was helpful!
Love it sir!
Glad to hear it! Thanks!
Very useful video, thank you very much
Glad it was helpful!
This is a really great video! i'm just starting out. Thanks for the ideas🙏🙏
You're so welcome! I have a number of other Midjourney videos on this channel, and I'm working on a detailed Stable Diffusion tutorial as well.
@@hml just joined your discord. Definitely gonna binge watch your videos. AI seems to be the future😅👍🏻🙏🏻
Thank you very much!
You're welcome!
So helpful --Legend --thanks
You're welcome! More videos (including Stable Diffusion) are coming soon.
I was on the feed when you did this
That's awesome! To be fair, I did it a few times while preparing this video, but it's still cool!
So cool thanks
Glad you like it!
thaaaaank uu!
Thank you!!!
every time i hear your discord notification sound i think it's mine. lol. Thank you for sharing your knowledge with us.
Yeah, I was getting that as well when I was editing the video :-) It's my absolute pleasure!
Hueregeil! Wirklich hilfreich, danke!
I have a feeling that Google Translate doesn't translate the first word properly 😂
@@hml 🤣 that's the Swiss word for "mega cool"! But google, i guess, is too "literal" in translating it. 😬 Anyway, great channel! Learned a lot of things never knew before.
Thank you 👌
Welcome 😊
Wow, that a great informative video! I have been trying to do this for ages in Midjourney. I am actually trying to do something similar to what you showed, but with an animal. I am not having a ton of luck getting consistent results, any suggestions?
I did not try it with animals, but I really don't see why it wouldn't work if done right, since it worked for me even with random shapes. Show me what you're trying to do in the Discord (@davidtab), and I'll try to help if I can.
Hi! Thank you this was very helpful :) I have a couple of questions tho. When u Click on the emoji button to send yourself a message with the seed of the picture, did you have to set up anything to be able to do that? because I don't have that envelope button when I open my menu there. The second question is, when you upload your picture and then want to copy the link, did you do anything for the last three options to show up? because on my page the last option that shows is to delete the picture (in red), so basically I don't have the save or copy link options. Please help
HI! Sure, you're very much welcome! About the envelope - no, you just hover your mouse over the Midjourney response, and click the "ADd reaction" button, and you can type in "envelope" in the search the first time you open it if you don't see it in the popup that appears. About copying the URL - if you're in your browser, you can click on the image, it will open up in a popup, then right click the image and choose "Copy image address". That should work. But overall, it's easier to do in the Discord app.
Great video! I'm eager to try incorporating AI into my videos on the channel! Surely will give it a try! Thanks for the tips!
Glad it was helpful! A few other videos on the subject coming soon.
@@hml this is amazing news! Looking forward and subscribed 😀
Amazing thx🔥
Glad you like it!
Genial, gracias amigo!
Glad you liked it! More are coming soon!
Thank you!
You're welcome!
I'm a PhD student from chemistry. In addition to doing experiment, we also need to publish academic papers which contain graphical abstract n other illustrations. Sometimes, if editors are interested with our work, they will also invite us to design journal covers. However, many researchers are not good at using software for graphical design, n we will subcontract the job to third-part companies whose offering prices r really high n will take 1-2 weeks . So i think the AI can be used in scientific drawing on different types of pics, such as graphical abstract, cover design, illustrations (like hieratical pie charts in review papers). I hope u can make a vids on scientific drawing for us if possible, thank u!
I'd love to do that! Maybe if you can show me a few examples of how those illustrations look like, I could make a video about that? ping me on Discord (davidtab#7572) and send me a few examples, and I'll see if it's something I could do.
@@hml thank u for the response. could u gimme ur e-mail, then i can send u examples? still don't have an Discord but i can register one
@@user-js9ji3ym1h sure, you can use the email from the "about" section of the channel. Press "View email address"
GREAT VIDEO, GREAT MATERIAL. CONGRATULATIONS FROM SPAIN. TAHNKS FOR ALL INFO. OLEEEEEEEEEEEEEE!!!!!!!!!!!!!! TORERO
Thanks for watching! I absolutely love Spain! OLEEEEE!
Thank you! 💗 there is a way to create the image with the style of the image already I have ? ( Promt + style of the image I have) If you can, let me know the method to do it .
Sure, you can do a img2prompt, similar to how I show in the video, and then modify the resulting prompt to generate what you want it to show.
I'm really interested how your custom generated images with your picture as a template turned out that good and realistic. When I try to use some photo as the template (and I used them a lot) generated picture doesn't even resemble the face on the original image - completely different persons and faces. Any ideas?
facing the same problem
@@thelovewins0126 I have tried Image Weight parameters to the extent, but to no avail. As I read on Reddit, MJ purposely alters the faces to avoid deep fake law suites.
same here would really like to know how !!
MJ doesn't modify any faces, as you can see it's easy to generate images of the pope or Donald Trump - if anything, they would block those first. Image Weight isn't supported in Midjourney v4, but it is supported in v5, and I found v5 is better at image 2 image than v4.
The paid version isn't better than the free one in any way, just that it can generate more images... However, MJ did discontinue the free version, unfortunately, so as far as free image generators go - the best options now are Blue Willow and Stable Diffusion.
You're awesome man
You rock!
....and the best part of this journey with AI is that we are just getting started!
Absolutely! There are so many things already that moved forward, and that's just the beginning. This technology is only 5 years old, and look at the progress it already did!
Surely your channel will grow a lot, compared to everyone I've seen so far your teaching was perfect, keep improving that soon it will be the most referential of all, just missing the instagram address.
Thanks a lot! I'm glad the explanations were clear. I actually don't use Facebook/Instagram, and even made a video about the reasons why.
Your information is pure gold. Do you know if they killed the ability to give the uploaded image weight? Here is the error I get (Invalid parameter
Cannot have custom image weights with --version 4) So, it could be me, but I just used --IW command. Thoughts?
Unfortunately, the --iw parameter isn't supported in V4 of Midjourney, only in previous versions.
for subscription, does it support the cancel anytime method? just in case I didn't wish to continue the subscription.
yeah, you can cancel at any time. Just one click away.
great video 👍👍👍👍
Thank you 👍
Midjourney is indeed an amazing tool that can quickly turn your ideas into stunning images. I've used it before and was impressed with how easy it was to use, and the results were fantastic. However, I'd also like to recommend Blue Willow, which is another great tool for creating stunning images quickly and easily. It's definitely worth checking out if you're interested in exploring other AI
Yeah, I will release a new Blue Willow guide soon as well. I do think Blue Willow anywhere near as good as Midjourney, but it's definitely good, and getting better, and worth mentioning.
when I copy the link and paste it and addedd my words... the face in the results are different from the uploaded image. But in the tutorial you are getting similar kind of faces. What I'm missing..?
It's hard without seeing the actual results - are you on the HML Discord? Tag me there, and I'll try to see what you're doing, and maybe give you a few tips.
any prompts to create an avatar of your own image, only with face in foreground? I'm trying everything but different images from me come out
Watch my video on Midjourney Face Swap, I show there how to fix it.
hi , most youtuber did not mention how to create your own pics avatar to get real or very close feature from the same photo you shoes as source neither you , so what is the keyword you use to get at least 60 % or more real you or me avatars , have a good rest of day. thanks
Hi, it's less about the prompt, and more about the image - Midjourney doesn't copy the image, but rather "gets inspired" by it. So you need to give it the proper inspiration - preferably an image with as little background as possible, where you look straight into the camera work best. But definitely try different images, and see what works best.
I already have images of my architecture project, would just be nice to have an Ai tool to add foliage and textures without having the painstaking hours in photoshop - hopefully soon!
Should be relatively easy in Stable Diffusion with ControlNet. Midjourney doesn't give you enough control to do something like that.
Excellent video. Thank you. One question, I upload an image of my face and prompt MidJourney to use the image of my face and put it on a superhero for example. I can not get it to work. It always renders the image as just my face size still without ever showing the entire superhero body with my face on it. I have tried all the suggestions of show full body, full body perspective, wide perspective as well as specifying what the shoes and hands look like, but I always only get a face shot at best. Any suggestions from anyone? Thanks!
Usually names of shots, or camera leneses work best, so try to add "wide shot", or "24mm" to the prompt
big love
Thank you so much :-)
Hey man great tutorial, buy I didn't understand the part when you said if my face changes, what to do when my face is not the same as the picture
Generally any small detail in your original photo gets exaggerated. So for best results, use a straight face, with minimal background, otherwise the background/angle will influence the way the image is generated.
What AI can I use to recreate a finished painting I did but I want it to come out looking like an anime? I painted Batman on top of a rampart of a castle but I would like it to look anime, and I don't do anime. Can it be done?
Yeah, it probably can be done. Can you share it with me on the Discord, and I'll try to figure it out?
Thank u
You're welcome!
i also do lot of experiments with own images and i also found out that it does not create hands well. Have you ever experienced addtng another charaacter in your photo? I wanted to add myself with a celebrity but the issue is that some of the text fits on me and some on celebrity. For example if i want to add earings on female celebrity it add on me. If you have the solution of this particular problem please guide. Also let me know what to add for realistic photos?
Unfortunately, Midjourney isn't a very good tool for this. In Stable Diffusion you can add "masking", that can mark what to change, and what not. So that would be a better option for you. I might make a video in the future that covers that.
Excelente video. I've been struggling to get images with closer look from a picture of some including a prompt. The result looks very different from the original looks. Can you give me an advise? Thank you very much for your content.
Glad you liked the video! If you'll use Midjourney on our discord, I'll try to see what you're doing with the prompt, and will give you some tips. But as a pointer - the cleaner the original image, the closer the result it is to it, so always keep that in mind. Also, sometimes it helps to overlay the image over the result again.
@oswaldosaid611 I recently experimented with a different AI called Blue Willow, and I'm really surprised by the results. You can try this with BW
Yeah, we have Blue Willow setup in our discord as well, but it's still far from what Midjourney, or properly setup Stable Diffusion can do.
Hey I have question,
How come when you use reference image your actual face remains very close, when I am trying to edit with reference image, entire face and body structure get changed.
It's hard for me to help you without seeing what exactly you're doing, however I almost always in the HML Discord, so come show me, and I will try to help you.
could you use these ai / avatars as 3d averters for videos
Not these ones, but stay tuned for my next videos!
If I have a picture on a white background, e.g. fruits - can I ask this program to paste these fruits into some nice kitchen and make a realistic visualization?
You cannot do it well with Midjourney, but it's easy to do with Stable Diffusion, and a tutorial is coming soon.
hi there. i would like to ask regarding why the ai always generating other peoples faces instead of mine. i did upload my photo to midjourney and follow your /imagine prompt mypic.jpg working secret bunker?. Thanks in advance
It's hard to say without seeing the results you're getting. What is your username on Discord?
NICE
Thank you!
Hi i am unable to get the copy image link option. I only get the copy message option. Is it because i am on trial version?
No, @ShabbirAliKanchwala, it depends on how you use Discord - in the video I use Discord App for the desktop. If you run it in the browser, you need to click on the image, and copy the URL from there. If you're on mobile that's a bit more complicated. If you joined the Hacking Modern Life Discord, on the #midjourney channel there are 2 short videos that show how to do it on desktop in the browser, and on mobile. Try to watch this, or ask me in the #midjourney-questions channel, and I'll help you.