How to Combine Images in Midjourney
Вставка
- Опубліковано 15 чер 2024
- 📙 Midjourney COURSE mastersofmidjourney.com Beginners to Advanced
🚀 FREE Midjourney Cheat Sheet tokenizedhq.com/freebies/mj-c...
🔗 FREE Promptalot Extension promptalot.com/extension
🔗 FREE Supporting Material tokenizedhq.com/freebies/vide...
Folder: 2023-12-14 - Combine Images
🌐 Check out the full blog post:
tokenizedhq.com/midjourney-co...
-
📰 AI Newsletter 👉 Your Inbox tokenizedhq.com/newsletters/ai
🐤 Follow me on Twitter / chrisheidorn
💼 Follow me on LinkedIn / christianheidorn
📸 Follow me on Instagram / christianheidorn
💬 Join the Tokenized AI Discord tokenizedhq.com/invite/discord/
📺 Recommended Related Videos & Playlists 📺
Monetizing AI: • Make Money with Midjou...
Midjourney CHARACTER Design: • Create CONSISTENT Char...
Activate GOD MODE in MJ: • Become a GOD in Midjou...
The TRUTH about Prompts in MJ: • The Surprising TRUTH a...
Create CONSISTENT Characters in MJ: • Create CONSISTENT Char...
How to Use --SEED in MJ: • How to Use the --SEED ...
Ho to Use NEGATIVE Prompts in MJ: • How to Use Negative Pr...
Create LOGOS in MJ: • Midjourney Logo Design...
-
Want to know how to combine images in Midjourney?
Until recently one of the major weaknesses of Midjourney was the fact that you couldn't easily build on top of an existing image to create something new.
While image prompts served a certain purpose and could be useful, they didn't quite meet the expectations of what users wanted to achieve.
In this guide, I'll show how Midjourney v4 opens up an entirely new world of possibilities and how you can easily combine images in Midjourney.
-
⏰ Timestamps ⏰
00:00 Introduction
00:55 Switch to Midjourney v4
01:20 Upload Images to Midjourney
01:59 Example 1: Combining 2 Images
03:31 Example 2: Adding Styles to the Prompt
05:35 Example 3: Describing a Scene
07:18 Example 4: Using Faces of People
09:41 Example 5: Describing another Scene
11:35 Example 6: Remixing an Image Variation
12:46 Example 7: Why this doesn't work in Version 3
14:14 Combining more than 2 Images
15:23 Final Words
-
Credits for images used in the tutorial:
- Image of Shaolin Warrior by Wade McMaster: www.pngfind.com/mpng/hJmwoRw_...
- Image of Mountain Castle by Jorge Jacinto: / 48273027244811404
- Photo of Abraham Lincoln by Alexander Gardner: www.loc.gov/pictures/resource...
- Image of Viking by Unknown: www.pixel4k.com/viking-warrio...
-
Music from #Uppbeat (free for Creators!):
uppbeat.io/t/ilya-kuznetsov/dawn
License code: P7ZI0TKVQ0DWXMH9
-
#midjourney #midjourneyai #midjourneyart #aiart - Наука та технологія
I cant believe this was only 4 months ago. Crazy how the software, and you, are evolving so fast, together. Thank you for sharing your knowledge.
It's shocking how bad this video actually was 🤦🏻♂️🤣
You are super consistent I was kind of hesitant to subscribe because I have this loyalty thing where if I have someone I'm listening to about a subject I usually stick to just that one ( I know it's stupid ), but you convinced me. Nice work.
That's really nice to hear :)
Brilliant Tutorial! your content on Midjourney is well explained and to the point!!! thanks a lot!!!
Comments like these are humbling ☺️ So glad to see what people appreciate tutorials that go into more depth.
Thank you for showing this. This helps me a lot
Super interesting and helpful! Subscribed, looking forward to more.
Thanks. Appreciate it!
Thank you Chris, this was awesome.
a real big help, excellent explanation. 👍🐺👍
No problem 👍
very well conveyed and very interesting - have subscribed, background music doesn't bother me, I find it soothing , has a hypnotic touch
Thanks. I agree about the music. I think I'll try to figure out what more people think and potentially make some slight adjustments then.
Great content. Just something to consider. I know it doesn't bug everyone, but you will lose some people who are distracted by looping background music. I'm one of those people, but I'm not alone. You have an excellent presentation style, so while you are instructing, you just don't need it.
Thanks for the feedback. I didn't realize it bothers some people. Do you think it would help if I simply reduced the music even further? I just find it a bit too quiet when absolutely nothing it playing in the background.
No music while talking is my pref
@@TokenizedAI the thing is us binge watchers have to put tutorials on 1.5 speed, if there is music on it, sounds insane. But again you cant please everyone ;)
Yeah, I totally get it. I've actually removed the background music while I speak in all my latest videos.
Yes. Great video. Just
Remove distractions. Good luck with your channel.
Thank you professor! This was a really cool class. I learned a lot.
You're very welcome! 🙂
Super informative. I am so appreciative of the knowledge you are spreading. Thank you.
So nice of you! I'm working on more :)
Great done without any doubt, thanks for every minute. 🙏
Really good video ;-) keep up the good work :-) loved the part how you put the prompts as subs ;-)
Thanks! Great to hear the people are noticing these details 🙂
I’m learning so much from you😊👍🏽
Thank you! Thank you! Thank you!
You are so welcome!
Amazingly useful, thanks! 😍
No problem!
I've had some really good success with upscaling images I like, throwing them into Gimp to use the masking tool to remove or add things I like. I take the new image and than regenerate a prompt based off of my new picture. It's like poor mans Stable Diffusion, but the results are similar if you use weighting in your prompts
Yep, that works as well. I'm actually going to have some elements of that approach in the next video.
Great video, thank you.
Thanks so much!
great video! Really helped me understand how to upload images better to use as references.
Glad it was helpful!
very informative, thanks for sharing.
Thanks so much. It's comments like these that keep me motivated :)
Liked and subscribed! 😉
excellent vid
tks for sharing !
Thanks! Appreciate it 🙂
Thank you.
Thank you for your videos best on you tube! making my adventure Awesome!😤😊
Have fun! 😁
Excellent video.
Thanks so much.
very helpful video thank you
You're welcome! 🙂
great vid, thanks! quick tip: you can copy the image URL just by right-clicking the image in Discord and select 'Copy Link'
True, but then the URL contains a query string that reduces the dimensions of the image. If you want the full size original, you need to open the original. Of course, it may not matter much because the differences minimal. But I think it's good if understand the difference.
thanks for the tutorial
🙏🏻
very nice, thanks
Glad you enjoyed it 🙂
Wow, I've never heard "url" pronounced like this..very interesting 2:12
Hahaha, I guess it's one of those things where you either do it one way or the other. Saying "U-R-L" just feels cumbersome to me :)
Earl
Wait till you notice him say Abraham.
Seemingly perfect English but reading abstract words was never taught... Maybe he's a bot
@@janeblogs324 1:29 there it is. Amazing. Yes i suspect he's a bot too.
@@norzillian9867 @janeblogs324 You guys are killing me 😂
So, just to clarify. I'm German-born and German-based, but spent most of my childhood in Canada and France. I've lived in 3 different English-speaking countries (Canada, UK, and South Africa). Trust me when I say that being exposed to so many different accents takes it's toll on you and that's why my English sometimes sounds like familiar North American yet sometimes weird pronounciations sneak in.
Specifically in the case of "Abraham", I am very aware of how it would normally be pronounced. However, when I'm speaking in front of a camera, German influences will subconsiously get involved. The way I said "Abraham" was very close to how it would be pronounced in German. But yes, seeing it now, I agree that it feels out of place.
Either way, bot or not, maybe this helps provide some context as to why my English doesn't fit any of the typical categories that you'd expect 😉
For artist, it's good to seek ideas / fast visualization.. but when it doesn't go as I want, it's easier & faster to draw it myself..
That's true. In many cases, the inspiration is enough though.
Thanks
You're welcome :)
Which one of your videos tutorials shows how to combine 2 characters together, where they interact , talk or walk together in a conversation? ( as n image, not as a video)
You'll want to check out the one on Multi-Character Scenes. But I must warn you, it's still very difficult to create these reliably in MJ without using external editing tools.
I do think there is a difference if we put a prompt the very first or th every last based on the images generated when I switch the positions of the prompts. However, I do not know if this is just my random experience. Relatedly, much harder to tell if there is a difference weightage for the "middle prompts"2nd, 3rd, 4th...(so on) prompt position. Anyone can comment on this?
I think that's mostly randomness. If your prompt is just a single concept, then I do believe the first part of the prompt will probably be treated as more important. However, with weights I do not believe this to be the case. The results will ultimately look different simply because "house::1 red::1" and "red::1 house::1" are objectively different prompts. So even if you set a --seed, it can still tell that it's not a carbon copy of the first prompt. But once it's mapped within the AI model, it will still produce very similar results.
I am very happy that there is a specialized channel in Midjourney Useful information, I want to know is it possible to reach the stage of working in a specialized field in design through artificial intelligence Is it possible to reach this stage with training and we will no longer need design programs?
And I suggest you make a clip, how to avoid errors when designing in AI and with the same fingers, my brother always come results for a strange number of fingers 8 and above😅, and thanks bro🙏
I think it's still a bit early for that but I'm sure that many people will be able to offer specialized services that makes use of AI tools.
Wow this is so informative. You literally broke it down to the smallest detail. And showed how to 🎉
Thanks! Just trying my best to give everyone as much helpful advice as possible :)
Great video :D ! Can you please explain how do you change between versions V3 and V4 ? Thank you.
To change your default version you have to activate the one you want in the settings panel. Enter /settings to access it. Alternatively, you can also just specificy the version at the end of your prompt by entering "--v 4".
@@TokenizedAI Thank you ^^
Will you ever remake these using v5+? Great stuff, very helpful, instant sub!
Not sure to be honest. Combining images isn't really considered "new" anymore and it might feel repetitive for existing subscribers. Don't you think? 🤔
how can i create the effect used in the cover image of this video?
i am trying various prompts but still failed.
thank you
The cover image was created by simply entering the prompt "half robot half human". It's not a effect, it's just what I told Midjourney to create.
Thanks for this video!
Do you think that we could create pixel art sytled low resolution images? I am asking because I would love to use it to generate game assets (or at least some concept art that could be easy to made into pixel art by a beginner like me :))
Yep, just add "8-bit style", "16-bit video game" or "pixel art style".
Thanks for your videos! How does one generate a blended image of two faces that retains the look of each face, but is split down the middle, like a split personality scene/portrait? I recall seeing an example of that somewhere, but can't find out how to instruct MJ to do it...
"half human, half robot"
@@TokenizedAI Apologies; I don't quite understand. Say I have two MidJourney portrait renders of the same character, one with a big smile and the other screaming in anger; how can I create a combined image (with Midjourney) that shows a split down the middle with the happy face on the left and the angry face on the right? As I mentioned, I've seen this before, but don't know the actual consistent command(s) to do it.
Ah, sorry. I misunderstood. You can't really do that in MJ. You'll need to do those final touches in Photoshop
@@TokenizedAI I swear I've seen it before - through the public renderings when I first started. I do think there's a way to do it... if I find out, I'll be sure to share!
hi, your tutorial is awesome, that join subscription, question, if I want to create the logo and put a domain name like "House", can I create a house image with a title name to be shown exactly called "House"! The reason I asked is that every time I put the text name as the title, the result shows a different text name.
Midjourney is an image AI model, not a typography AI model. It's not capable of generating sensibel text. Whenever you see something in an image that resembles text, then it's only because the images Midjourney was trained on, often contain text. The AI doesn't even know that it is creating text-like content.
thank you so much for your response, I guess I would have to edit myself from other editor software, again, I watched all of your vide and learned lots from you. please keep going thanks.
Oh, I forgot to ask one thing about the art ai, if I want to create a header like using a full wide banner, is this possible can be done using Midjourney!
MJ isn't capable of creating just any random format you like. You'll still need an image editor. You might find this video helpful: ua-cam.com/video/pmY_ysnL5bQ/v-deo.html
These are helpful but do you take requests? I am able to pick up what I want to do when I watch multiple videos and that is good but I have some direct questions that could probably make decent videos, or not.
I read all comments. So if you drop your questions or topics here, then I'll add it to the list. Whether it's prioritized or not largely depends on how many people are asking the same question.
Can you add weight to images to sorta control the influence of each Image?
Image weight is currently not supported in v4.
when I choose to remix a result, I don't seem to have the option to add/change the text prompt...however I see from this video that it is possible. What am I missing? Thx
You probably never activated Remix mode. If it's not activated, then it only generates basic variations. Use /settings to activate it.
Thanks again for all your help! I have a very annoying problem in that Midjourney has simply stopped processing images at 78%. I get an error message along the lines of "Sorry! Could not complete the job!
The job has encountered an error. The developers have been notified.
rubber-stalling-imitate" It usually happens when I am /imagine 2 images. I ran "Clean My Mac" and reinstalled Discord and looked online to no avail. Thanks for any thoughts you may have on this....thx
Brian
All the time or just occasionally? It's an error on MJ's side, not yours. This happens to me from time to time, but not all the time.
@@TokenizedAI thanks for reply…. It did happen “all the time” for almost 2 hours until I called it a day. Maybe Midjourney glitches after hours of continual use.
In such cases, you should be checking the #status and #member-support channels in the official Midjourney Discord. You'll get much better answers to your questions than here on UA-cam because it was probably just one of the frequent outages.
Question: I'm trying to change out the iris in an eye with another object. I've tried so many word prompt combo's but it's not working. Is this just a limitation of the program or do you know of a prompt that will let me do this? Thanks, and love the content!
Are you doing this on a extremely close camera angle or is the scene bigger?
@@TokenizedAI I've tried up close and far away. I am trying to replace/overlay the iris with a double helix. I've even tried inserting the reflection of the double helix in the iris but it puts the helix in the background (I don't use the word background in my prompt) or overlays it around the eye somewhere.
@@aaronsmith600 What prompt are you using? Or are you trying to achieve this by blending image prompts? That might prove difficult to do, to be honest.
@@TokenizedAI I have tried "a gorilla staring at the camera with a reflection of a double helix in its eyes" "Gorilla eyes looking at the camera and the gorilla's iris are replaced by a double helix" those kinds of combinations. I'd even tried "a gorilla sitting and holding a double helix in its palm" I tried to keep it simple before trying to narrow it down. I had to take a break because I was getting frustrated lol
@@aaronsmith600 Hmm, have you tried using multiprompts with text weights instead? I think you might be able to blend the concepts that way rather than trying to stuff 2 relatively unrelated concepts into a single prompt.
how do you upload image to discord, or do you just need the link?
Link is enough. But if you want to upload in Discord, check this: ua-cam.com/video/gTFUak-K6ws/v-deo.html
How do you take a picture of yourself, and simply make it more cartoonish or artistic? I can’t find anything that shows how to do this anytime I upload a picture I just use it as a reference but makes a new person.
You mean you want to essentially "import" yourself into Midjourney and reliably create images with your own character? I think the main challenge is that you would have to use a large number of images of yourself from various different angles. 1 or 2 photos probably won't cut it. But I'll try to dig deeper into this and see what I can figure out.
I have managed to do this a couple of times. It just takes a large number of re-rolls. I used two images of the person: one was a full-body cut-out with the background removed. I went to one of those background-remover websites for that; I think it's called "Remove BG". I had one of their entire body with the background cropped out, and then I did one of just the person's head from the same photo. I specified both of these images as image prompts. Then I went to the animation mode of Midjourney; I forget what it's called in the settings; it's not V4, it's the one for anime. As soon as I got an image that was actually a cartoon, I put that one in the image prompts. And from then on, it's a process of taking out image generations that aren't as good and putting ones that are better in and continuing to re-roll. The best advice I can give you for this is to lower the quality settings so you don't waste your money. I wish I could show off the results.
@@Jumpyfoot This is excellent input. Thanks for sharing. I'll definitely try experimenting with this.
@@TokenizedAI You're welcome. Another thing you could try is crop out just your head, say, and then feed that into Dolly 2. You could use Dolly 2 to make an image generation with a prompt saying something like: a king wearing a crown sitting on a throne; it will generate that for you, but in photorealism, right? Then, what you do after that is you shove that image into Midjourney, possibly with another animated image reference to go alongside it of a castle or a throne room that's animated so it decides to pick up on that cartoon style.
the image screen is not to sharp ...what you used ?
My newer videos have sharper resolutions.
Will be interesting if you also tried switching the order of the images
Order doesn't really matter.
@@TokenizedAI yes I just tried myself
can you create photo realistic image with 2 characters and assign specifics to each character? say man standing far to the left in the back ground wearing t-shirt and jeans. woman standing closer to camera centered, wearing a dress? I have not been able to achieve this. at best it puts 2 characters together. Blending didn;t help either. It omits 1 character or produces some disfigured features.
It gets very difficult to assign those details as soon as there are 2 or more characters
wow , interesting work flow thanks for the tut, can u tell me how you add mj to your server, thanks
Working on it as we speak :)
@@TokenizedAI i figure it out 😀, also just finished my story with chatgtp it has come out really good
Ok.. stupid question.. I uploaded my images to discord, then as soon as I started typing /imagine.. all of the images disappeared. Now I can't find them. Where are they being stored?
They disappeared? Sounds like Discord may have filtered them out because their image recognition system detected something they don't like. That usually why that happens.
What I'd like to do is to add a character to another character. I created a landscape, and on it I want to have the two characters I created, not mixed them. Is this possible?
You should watch my character design series.
@@TokenizedAI I'll check them out.
“Earl”
🙈🤷🏻♂️😂
Late to the discussion; can I use your instructions in V5 or v5.1 ?
The logic hasn't changed.
Great work 👍 just one question.. If I wanted Abraham Lincoln, and the samurai in the same new image, how can I do that? In midjourney v5
You mean both separate characters? That not what image blending does. It will always try to "blend" them together. Hence the name.
@@TokenizedAI So there's no way to do that.. 😑 Thanks 👊👍
At least no simple way that I can think of, and certainly not by simply blending to unrelated images.
@@TokenizedAI thanks for the clarification. I was going nuts already 😂
how can you let MJ know what image you want the style to look like and what one the object
I'm not sure I understand the question. A style is applied to an image, not the other way around.
@@TokenizedAI what im trying to say is if you give it 2 images they will both have a style to them so is there anyway to let it know what image you want to influence the overall style of final result?
Wonderful presentation. I am new to this whole Ai art. Can anyone recommend software for beginners? I have something specific in mind to create or have someone create it for me. Thanks
Midjourney is a good, relatively user-friendly start with their free trial.
@@TokenizedAI thank you so very much.
Are the reference files stored somewhere?
No. These were just for demonstration purposes. I recommend you test this with your own selection images.
I've watched the video a dozen times now and I still can not get this to work. You do something at 1:37 and mines not behaving like that. I keep getting the message that Discord won't use a single image prompt! Cheers for the video
Are you referring to the actual "uploading" part or when you enter the image prompts? Can you share the exact error message that you are given?
I have finally managed to get it working. I wasn’t leaving a space after the first prompt! Cheers Rob
You left out the critical first step of creating the URLs. The video begins with the comment, I did some prep work... and that covers that information. Computer commands don't like missing steps
The prep work had nothing to do with anything technical. I don't think I need to explain to people how to use Google 😅
How may I use my actual logo to be printed in a car, shirt, cup, building...?
You'll most likely need to use an image editor for that. AI models aren't really built to do what you are trying to do.
Will you update this with V.5?
The process hasn't really changed though 🤷♂️
Hello friend, make a tutorial how to turn photo with actor into iron man or batman for example)
Noted. I'll see when I can fit this into my video schedule.
First person I've ever seen call URL's "Earls" 🤣🤣🤣 Great lesson nonetheless!
It's a thing. The majority says U-R-L and I've started doing the same just to avoid the constant comments about it. But yeah, a lot of people also say "Earl". It's one of those endless debates.
can I add weights to image links?
In v3 yes, though that's done via the --iw parameter. v4 doesn't support image weight yet.
@@TokenizedAI and I would apply this " --iw " at the end or after the link? sorry newbie here.
Check out this article: tokenizedhq.com/midjourney-image-weight/
But I should point out that image prompts work differently in v3 than in v4. The article explains it.
Hey, could you help me please
i make midjourney and pay, but i wanna make it take my pic and another pic of my friend then make us beside us on some place
Cannot be done in a single prompt or without other tools.
How u switch to version 4?
Version 4 is now the default version. But you can invoke it either by using "--v 4" or by activating it via /settings
How do I apply these on mobile?
Discord works the exact same way on mobile as on desktop. Nothing you need to change really.
Interesting. Thanks for the demo. Though if that's not your own artwork you uploaded, or didn't buy them, it would be cool to give credit to the artists whose work you're starting with if you're encouraging other people to do the same methods. Ideally you'd get consent if publicly demonstrating with their work so it's not a surprise if they stumble upon this video. A lot of artists on Artstation are upset about their work being used without consent. And if anyone says "it ends up completely different" - it wouldn't look like that without that original input. I use MidJourney myself, ( I'm not anti a.i.) just pointing out ethics, especially if you're making money off views and ads.
You make a very valid point! Thanks for bringing this up. I totally agree with your statement. I'm gonna add those credits asap.
@@TokenizedAI at least you're not against crediting people
Some AI users completely dismiss it
Why don't you press Ctrl-+ on Windows or Cmd-+ on the Mac a few times before you start recording, or record at a lower screen res? Most of your screen space is wasted and all text is tiny.
This was only my second video. Please excuse the lack of professionalism during my baby steps.
@@TokenizedAI The content is interesting. I just wanted to point out that inside Discord you can use Ctrl +/- to make the font size larger/smaller (I think, at least on the Mac it works), and increasing the font size inside Discord will help viewers greatly, and it's very easy to do! Keep them coming!
@@AdamTwardoch Got it 👍🏻 I think I may have a better solution for this that doesn't require increasing the font directly.
good but I expected more complex jobs like the one you show on the cover of the video
The one on the cover was the simplest one 🤣🤣 "half human half machine"
Don't assume that a very detailed image has to be the result of a super complex prompt. That's part of the problem. I swear, so many of you are overthinking this stuff 😉
Why aren't you using the text weights that you said were so important?
First of all, because the video is about blending 2 images prompts. Secondly, because there's a time and place for weights but that doesn't mean that you always have to use them. There are no hard rules.
👍
🙏🏻
Is there a free model that does this ?
Not sure
Don't worry, viking abraham lincoln doesn't exist.
Viking abraham lincoln :
😂
why am not able to open ur site?
Some locations are blocked because the site receives extremely too many hacking attempts from those locations. Sorry. 😔
sad to know. But Bangladesh is free of hacking stigma.
You can also just use imgur instead of cluttering up your discord
Since I run MJ on my own server, I have a dedicated channel just for uploads. Much more convenient than Imgur to be honest.
@@TokenizedAI as do I, to each their own though.
👋
go opensource and use free/open stable diffusion tools, don't be subjects
I already use Stable Diffusion as well. But I must say that it's not nearly as accessible as Midjourney for the average user. And if you don't install it locally (which most people aren't capable of, unless they use DiffusionBee), you're still a subject anyway.
@@TokenizedAI you are right, there is a tradeoff in practicality, but in the end one is free and the other will cost fortunes as digital artist become dependent on updates $$$. Also consider that you can finetune stable diffusion much more and there are tons of software being developed by the open source community and academia around stable diffusion.
I don't disagree with you 🙂 I'm just saying that in the end the majority will still pick convenience.
Why are you even bothering with commas? I thought you said in another video that punctuation didn't really mean anything to the AI.
Because I find it easier to read. It's not for the AI, it's for myself.
I love your stuff but PLEASE say YOU ARE ELL not Earl for URL :D :D love you
Hahaha! Yes, I've started saying U-R-L now because everyone's been bullying me because of that one video XD
@@TokenizedAI hahaha yea internet peep can be cringe sometimes but some rules can't be broken 🤣 but you do you. You're dropping value and that's all that matters really ☺️
I don't take any of it personally. I just find it amusing how it irritates people. I know 80% say U-R-L. For context, saying "Earl" is more common in some other languages, which I believe is why I may have adopted this pronounciation over time.
@@TokenizedAI gif is another one. If you call it Jiff I have to unsubscribe 🤣🤣🤣
@@shrvn110 Uh oh..... 😂
You skipped a main part. You didn't say HOW to get the link. The URL doesn't just show up. I had to right click the image then open in browser first, then get the URL. Beginners to me means EVERY step just in case.... IJS
Well, you figured it out, didn't you? 😁 Not such a beginner afterall.
Your content is great (so thank you for that) but I entirely lose the ability to focus every time you say "earl." It's an acronym and the letters are pronounced independently - "U R L"
I wish people would just focus on what matters and not blow these totally insignificant details out of proportion. It just doesn't matter unless your OCD overcomes you.
Please, lower the volume of your intro and outro music. It is glaringly loud.
This my only my second video. Can you kindly confirm it's no longer as bad on the latest video. For some reason I don't have this issue with my speakers.
@@TokenizedAI I just watched your latest video "Become a GOD in Midjourney with Text Weights", and it is still an issue, even though not as bad as this video. Your music levels should match your voice levels. My advice (you really do not need speakers for this), in the video editing program you are using, check the decibel levels and make sure that it matches your voice.
@@sun-eye Hmm, I think the editor I'm using might be too "simple" for that. I guess I'll have to see whether there's any way I can normalize the audio after editing.
Thanks for the advice. Appreciate it!
@@TokenizedAI You really should because it is the little things that count. Especially, your content is really, really, good. I am actually binge-watching more of your videos, learning a lot. Unfortunately, I already know that when the intro or outro is about to start, I have to lower the volume a little. LOL!
😂🤦🏻♂️
Demantébiasjiomas
Supercalifragilisticexpialidocious
this is not for a beginner like me, so confusing
I'm curious. Which part to you find most confusing?
abraham link colon
🤣
Don't teach how to use stolen works...
I'm a children's book illustrator. This is no the way the pages layout ahould be designed.
The storyboard is really important. You need to design the whole book to see the story flow.
This is a collage of images.
The perspectives should be consistent on all the characters and also the lighting.
The shadows are not correct in the video.
Sorry, but designing a children's book require a lot of study and learning. Also photoshop.
Sorry, I can't quite follow. How is any of what you just said related to this video you just commented on? The video has nothing to do with children's books.
Or you know, you could just practice drawing
I understand that change is difficult. Artists don't want to see this. But let's face it, this stuff is only going to increase. Rather be among the first out of the gate than get stuck in the past 🤷🏻♂️🙂
We do what we enjoy? If you enjoy drawing keep drawing.
This. Just let people do what they enjoy. Not sure why everything needs to ve turned into a clash of ideologies.
@@TokenizedAI If you actually look at the amount of work that goes to the artworks AI steal from, then you would understand
It's the hard work of millions of people being stolen, and as you say, "being left in the past"
That's just insulting
@@nx9919 The only thing I find insulting is how little mental power is being spent on actually understand what is happening here. Instead, overly simplistic labels are being used to express initial frustration and fears, whether justified or not.
When an AI is trained on content, it's not storing a copy of that exact work somewhere in a database. It is being trained, which is not fundamentally different from someone who looks at other people's work for inspiration or who starts his career learning how to mimic other artist's styles. And let's not pretend that's not how many artists learn their craft. The fact that an AI can do this much quicker and more efficiently than a human doesn't make it worse. If that were your standard, then you'd have to make factories that use any form of automation illegal.
Before anyone can have a sane discussion about this, it's important that everyone understands what is actually happening here from a technical standpoint. Just because something "feels" like stealing, to you doesn't mean that it is in fact "stealing". Especially when assessed within the legal frameworks our societies operate under. That's what the court of law is for.
I understand that people are upset. It's a knee-jerk reaction. But that doesn't mean that I agree with their "doom and gloom" views. This has happened many times over with other industries for various reasons. The outcry was huge...every single time. Why? Because an interest group feels threatened by an advancement in technology. There was no legal framework for it. SoiInterest groups battled. And eventually the law adapted to the new facts and a compromise was found which represents the new status quo that those very same people no longer question. It's a process of change.
You're entitled to your views. So am I. We do not have to agree. But neither of us gets to decide what is or is not legal when even lawyers state that it is completely uncharted territory.
One fact remains though. The technology is not going away. It's here to stay in some shape or form. Whether you like it or not.
man midjourneys optics are awful though, not their art but the layout
To be fair, I think this will eventually change. Releasing it on Discord was a smart move because it's easier and quicker to do. I'm sure a proper web app will come eventually.
@@TokenizedAI oh I know it will change eventually. I guess I started using the Dream by wombo app and it's so streamlined and doesn't show all these codes.
It's better than anything I could create , it's just a lot of the coding optics can get confusing and it's easy to get lost
@@djphlange Have you tried working with a Prompt Builder? Might make things easier for you since it gives you an interface.
@@TokenizedAI mmm i dont think so , ive tried several programs but ive had some decent success with what im doing with it just on the Dream app.
ill take those photos and then edit them or add motion in photoshop but thats about it
@@djphlange I recommend giving Noonshot's prompt builder a try. It's really useful as a guide if command line interfaces aren't your thing.
stop stealing others artwork
The way you phrase your accusation shows that you don't actually understand what a diffusion model does. I recommend reading up on it. It will help you make more convincing arguments next time you try to debate. These childish comments aren't helpful nor do they actually have any impact on anything or anyone.