How to Make an Animated AI Movie | Using Free & Easy Tools for Consistent Characters
Вставка
- Опубліковано 10 лип 2024
- This tutorial shows you a step-by-step workflow of how to animate your own stories using AI. We'll use Midjourney - or its free text-to-image alternative, Leonardi.ai - as well as a free image-to-video tool called Pika Labs.You'll learn to animate your own comic book characters, children's books, anime, AI films and more.
This animation tutorial is for beginners to advanced, teaches how to make consistent characters with AI art generators, and covers a new release of Kaiber AI's Motion tool. We'll also use a free tool called Canva to help us with our editing and animating.
AI FILM SCHOOL
🎥 Sign up to be notified when my courses are out at www.thereelrobot.com/
AI FILMMAKING ASSISTANT GPT:
➡️ chat.openai.com/g/g-hiKxJNAlp...
❌ Twitter
/ thereelrobot
00:00 Introduction
00:26 Image Generation Tools
00:47 Creating Characters
01:42 Canva (Image Editing)
02:33 Getting More Character Images
03:16 Easy Animation Workflow
03:49 Advanced Animation Workflow (Comics)
04:57 Animated Backgrounds
05:14 Effects & Comic Animations
06:19 CGI & Other Animation
07:09 Animation Styles
07:36 Kaiber AI - Motion
07:51 Outro
#aivideo #midjourney #aiart #future #aitools - Наука та технологія
I can’t believe how you packed so much information in an 8 minute video. It was amazing and clearly lays out how to create these videos. Thanks for all your hard work and videos.
I totally agree, was just thinking same thing myself! Can’t wait for the next video, this guy is awesome 😎
Appreciate you saying that. Thanks a lot for the encouragement!
@@TheReelRobotthis is amazing , im from brazilian congrats man. i dont know use promp ! sorry for english 1 more sub.i can use the same if one dog ? or animore animals ?
@@EstoicismenAccion I will write Portuguese and Spanish below for you:
Se entendi bem, você está se perguntando se pode usar um animal como personagem em vez de uma pessoa, certo? Sim. Na verdade, um cachorro é ainda MAIS FÁCIL para a IA entender. É menos complexo e mais fácil fazer com que pareça o mesmo cachorro. Os métodos seriam exatamente os mesmos que usei para a criança humana neste vídeo.
Si entendí correctamente, te preguntarás si puedes usar un animal como personaje en lugar de una persona, ¿verdad? Sí. En realidad, un perro es aún MÁS FÁCIL de entender para la IA. Es menos complejo y más fácil que parezca el mismo perro. Los métodos serían exactamente los mismos que usé para el niño humano en este video.
@@TheReelRobotthank you so much for this for digital designers ❤
This is crazy. Now anyone could make their own movies and shows for like $10 😳
You forget the amount of time and effort actually needed to actually make things. And most of the time you are rolling the dice and pray the AI give you something good.
@anhnhvn yeah, I get that. But what I am saying is, with the progression of technology, it's way cheaper than before, and in a few more years, AI will be more advanced and less of a hassle than it is now
This is such an informative and practical video! Amazing work, mate!
One of the best AI tutorial vids I've watched. Good job explaining.
Thank You!! Wow is all I can say. Was very clear and zero confusion. I just started on UA-cam so can't wait to keep on learning.
What a brilliant effing video. Clear, clean, and exactly what I needed to hear. Thanks, man 🙏🏾
You have officially changed my life with this video. Bless you, sir! This is what I have been trying to do for a week now. Thanks to you, I will be able to finish my project this weekend!
Wow, you explain it so well and logically. Thanks for sharing your magic.
Discord reduces midjourney image quality, instead click the the little web button under your upscaled image and download the image from there.
Definitely deserves a like and a sub 👍
You're totally right, it compresses a bit. I tend to truly download from midjourney.com/app when making videos, but I should be making it more clear to everyone in my tutorials
Thanks for this tip, bud.
Man, youre amazing! You've answered myone month problem with just 8 minute video. It's all i need. Tons of thanks!
Thanks! Let me know your next problem next time you encounter one, and I’ll see if I can address it
@@TheReelRobot Thank you!!! I'd love to explore my next curiosity, which involves adding two or more characters into a single panel and seeing if I can make them interact with each other.
@@kokolemu89 awesome! A prompt keyword that might help display two characters in the same frame at equal size is "two shot". Feel free to share your stuff with me when it's ready!
man, you're such an angel sharing this tut video 🥰
The best AI animated Tutorial Ever 🎉🎉
AMAZING! SO MUCH INFO, SO LITTLE TIME. TY
Some artist frown on their art being used for AI also AI cannot be copyright. I googled it because I really like AI as an artist it helps create references, and stuff like this is awesome. Just not sure how all the AI projects will monetize or play out in the future for original artists
Thank you so much! This is incredible!
Priceless. Thanks a bunch!
This was so amazing to watch!
This video just blew my mind.
amazing. I am very excited to explain this to my students. It is both easy, enjoyable, and awesome.
Well done man!
This is sick. Thankyou for the video 😻
great vid for sure!
Another great video. Love it ❤
I really like your delivery and style, excellent content and great tips. I save all your videos and rewind and replay so I can practice along with you. Great channel. Cheers bud 👍😃
💯💯💯
Thank you! Means a lot to get encouragement when your channel is new. Appreciate you taking the time to comment 🙏
@@jerzyjd They are very bad results. You have to learn to draw and animate like a professional if you want to dedicate yourself to it.
I'd love to see a workflow, in which you blend talking avatars (D-ID Studio, HeyGen etc.) with Animations (e.g. Pika Labs) and blend them into a conversation between two characters, using voices made with e.g. ElevenLabs. Cheers, brother! Thanks for the great video.
Great idea. I'm going to be playing around with dialogue-heavy stuff this weekend and see what I can do.
👍 *I DID NOT SKIP THE ADS. THANK YOU SO MUCH SIR. PEACE AND BLESSINGS TO YOU!*
This video was amazing and inspiring. Thank you so much
That was very informative. Thank you for sharing.
thk u so much!! your video so great
Bravo Sir!!! This is awesome!!!
Thank you for this content, it helped me a lot to get my ideas off paper
Also your videos are very informative and helps me a lot
Very interesting I watched half of the video and I like it. I was watching on my break, and I definitely bookmarked it to finish watching later.
Awesome to hear. Hope you come back!
Thank for sharing ❤
Awesome video, subscribed!!
SO EXCITING game changer is an understatement
Next level. Thank u!
Hi! I walked through this step by step but had issues with my copied image links to Mid Journey. None of the saved animated images on my web pages had PNG, JPEG, etc. at the end of the link address. I copied image links from my Google Photos, prime storage, and Wix website, and mid-journey doesn't recognize any of the links. Suggestions?
Wow Amazing Content This Video, keep the good work!
Thanks for sharing 😊
Thank you!
great info man thank you so much for taking the time to share you knowledge!
You are an expert techie guy. Keep rocking...
This is super awesome
Thank you so much!
Thanks! to you.. Yours are useful and helpful..
thank you man, can use this for my channel
This is helpful thank you
the fact that a video like this doesn't get the views it should really discourages me from doing anything else,the algorithm war is frustrating,great video
Thanks. It's because it's a new channel. Just over a month old. Even with good metrics, UA-cam takes a while to trust new channels
@@TheReelRobot maybe,i saw new channels performing since the start,maybe hashtag and seo?
@ghostproducer1988 tagging you to encourage you to release more stuff. 1 month out and it's at 35k views and growing faster each day. Algo just took a while to trust me
Thanks for sharing
thx so much!!!!
Great video! You'll have to give Augie a try sometime :)
good job
I tried this tutorial....it was fun😅
If you dont mind i would please love some more videos on how to do the comic strip , animation and cgi
Appreciate the input! I have gotten similar requests and will work on something. My next tutorial will be much more tips & tricks from the Top AI artists, but after that I will hopefully have time to focus more on a step-by-step animation guide
For voices in my animated videos my go to is Vocs AI
Hi just one question please, you got the character from Midjourney then to Canva then back to Midjourney,
can I put my own drawn by hand character into canva then back to Midjourney? or that won’t work ?
Ask away! Yes, it will work though the results are going to vary. So one thing you can do is drag and drop a photo of your hand drawings directly into the Midjourney discord --> Copy the Image Address --> paste the link to your prompt. It'll use your hand drawing as a reference photo.
If you're doing something with hand drawings in Canva first, you can export that image and drag & drop into Midjourney discord instead. One other tool you might want to check out for hand drawings: clipdrop.co/stable-doodle
Hi nice and easy cartoon animation tools tnx
thank you
You’re the best
Ones attending the funeral of their creative careers, unite!
People are still gonna want to see real artist this won’t last so don’t give up
🤝
It was nice while it lasted 😢
👌 👏 😍
This is awesome! Can you explain how you get the characters to be consistent in Midjourney? I know you copied your individual image links and added the expression in there, but mine generates new versions of the character even when I reference those. How’d you keep yours the same?
Have you been putting in "--iw 2" at the end of your prompt? You want the Image Weight parameter to be as high as it goes for consistency. If that doesn't help, try writing in "/prefer remix" instead of "/imagine", then after Remix mode is turned on, go to one of your images you like and do Vary Subtle. It'll let you put in a new prompt as you Vary Subtle that might help you get a new angle or appearance.
I'm releasing a tutorial soon on consistent characters using Everart.ai - that helps A LOT so you should signup for the waitlist.
Can you tell me where the Magic eraser is in Canva ? the search brought up No results, Thanks
Happy to. If you have a photo you want to erase part of already visible in Canva, click it to highlight it. Then at the menu near the top (a little left of Center) you’ll see an option to Edit Photo.
Once you hit Edit Photo you should see options appear in a menu on the left, and one should clearly state “Magic Eraser”.
If you still can’t find it, this should help: www.canva.com/features/magic-eraser/
I would like a 1-1 tutorial on this.
Hello. I want to upload my video on youtube using pika but there is watermark .. Do you have any idea to delete the watermark?..I really appreciate of you give me some solition.Thank you😊
Hi. It's a tricky one. What a lot of people will do is Crop the image, where they zoom in on the video just past the watermark. If you're using Midjourney to get your images, you can use the "Zoom Out" option on your photos before bringing them to Pika, and then when you crop it'll be as if you never even had to zoom in on the photo.
If you Google "video watermark remover" there's paid tools that can help too, though they might distort the image.
But the safest way to not anger Pika is to do some work that gets PIka's attention, at which point you can be invited to join the non-watermark creator program they have
I really loved this informative video . Too much useful video for earning money
Thanks! I am releasing a new video soon that's directly focused on making money with this skill.
Any idea when Pika is going to let you save videos without the water mark so they can be used for commercial purposes?
I’m not sure, but I would bet that you have to upgrade your plan. In the meantime, if you do something that gets on their radar in social media or discord, they might invite you to their non-watermark creator program. I’m thinking about doing more Pika stuff just for that purpose
Is it possible in Adobe animate cc
Rocking
I need your help with moving my animated image back to canva. It comes out with a black background, even though the background was erased prior. What should I do? Thank you.
So you mean that you get an animation from Pika or Runway, and when you bring it into Canva the background is black and you don't want a background in Canva, right? If so, Canva has a toolbar at the top. Click Edit Video --> Background Remover.
If you want to export it from Canva with no background at all after that though, it gets trickier. I use runwayml.com/green-screen/ to do that
@@TheReelRobot thank you! I will try it on in the coming week when I get some off days.
How did you make a video with a transparent background in Pika Labs? I just tried to create a video with a transparent picture and the video output didn't have a transparent background.
You can't make one with a transparent background in Pika Labs, but you can bring your video to runwayml.com/green-screen and remove it; or use Canva, Adobe, or video editing tools that allow for it.
thank you!@@TheReelRobot
tutorial bagus. Semangat👍
Terima kasih teman!
Posso te pedir p fazer isso com 2 desenhos que tenho?
Pq infelizmente não entendi nada
The best
❤❤
like informative
The Ia should solve this uderwather effect. Or size cloud effect.
This a lot of work for Animation. There are various types of animation software out there. You might consider doing paid promotions for software
❤❤❤❤wanderful
This video ❤
I tried the drag image to make facial expression in leonardo ai, it didn’t work. What should I do? 😢😢
On the left side panel of Leonardo AI Image Generation -> Scroll to the Bottom --> Look for "Image to Image" and drag/upload the image you want to use --> Write in a prompt at the Top-Middle of Leonardo, like "happy facial expression" --> play with the slider that says INIT STRENGTH until the image comes out the way you want it to
Please why can't I use pika labs? It keeps telling me to sign up and join the wait-list. After signing up, it says they will send me a feedback later. Why is that? Does it work on a mobile phone?
They’re launching a new version, which is what the wait list is for. But you can join the older version here: discord.com/invite/pika
2:45 what i should do if i dont have Midjourney ? Is there any alternative?
You should check out Leonardo.ai
They let you do an image upload (bottom left of screen) for characters. I hope to do a tutorial on doing consistent characters with other tools soon
Please sir, tell me the name of screen recoding,that you are using,, I like your videos,, keep it up,, weldone job,, here in Africa we are proud of you,,we are ready to be your student ,sir thanks
Hello to Africa! You mean the screen recording for the clips in the tutorial?
I actually just use the default Mac recording tool, and narrate over the video later in Final Cut Pro. I just edit the clips to fit in time with what I’m saying
God this is disappointing as an artist. I’m done
Don’t think that way. It’s true that the way you do work is going to change a lot, but it’s also true that you can create your own anime movie from end to end. That’s never been possible.
Adopt it, use the edge you have over everyone else as a real artist prior to these tools existing, and you can create a huge body of creative work
I know it's ridiculous, The only reason no one really cares right now is because it's not affecting their job, livelihoods.
They'll see once big corporations start going out of business because of AI.
yes
This is garbage
@@TheReelRobotstop lying. We all know we're gonna end up jobless because of freaks like you, until regulations are gonna be released.
at first when u were talking i thought its ai and freaked out 😭😭
Pretty reasonable thought given that I’m using adobe podcast enhancer which can get warbly, and then using freaky transitions.
But would definitely have added muscles if I was using an AI avatar
La video est top et bien expliquée. Mais, pour le moment, je ne trouve pas que l'animation avec AI soit assez avancé pour l'utiliser dans des projets. Peut-être dans un an ou deux.
C’est vrai, mais ça avance vite. Runway Gen 2 n'a même pas 1 an et s'est amélioré si rapidement.
Great stuff. However im struggling to to create effective prompts for action poses like kicks, punches, running etc. any advice anyone?
It’s very difficult, honestly. But at about 8: 26 in this video I have some tips: ua-cam.com/video/oP2me8Fc9Iw/v-deo.htmlsi=2G5x_44wuYi0D_gx
You kind of need to leverage photoshop or Canva to manipulate what’s in the shot. Or really spend a long time generating.
Pika Labs’ new text-to-video model is pretty decent for action, but you lose control over the consistency a good bit with text to video.
can I use a picture that was created in ai and put it into midjournery or Leo ai
Absolutely. In Midjourney you drag and drop the picture into Discord. In Leonardo there’s a place to upload an image near the bottom left
@@TheReelRobot thank you. If you could do a short on it that would be so awesome 🙏🏽
How to combine two characters? Say we have 2 people in 2 different photos and we want to generate a new photo where they hug. How to do that? Cuz the app would need both picture of the characters and then to combine them in a certain way
There’s a short film I made on here called missing parts where I did this:
ua-cam.com/video/p2QYda-CL8I/v-deo.html
The woman and the dog are in some of the same scenes together as a two shot, but what I actually did is use Canva to edit the dog into the image, and brought it to Runway to animate after.
You could also edit in the second character with Canva or photoshop, and take it to midjourney as a reference image using this video’s technique, and then use “two shot of [describe both characters]” and “-iw 2” at the end of your prompt to try and get Midjourney to give you an image with both.
If you get two characters in a shot, but one character looks off, you can try Vary Region on the character’s face, and then copy/paste the image address into the Vary Region prompt.
Some people use Facefusion, InsightFaceSwap, and other face swapping tools to accomplish this too.
But ultimately, what you’re trying to do is very hard. I tend not to do it myself often because it’s very time-consuming, and lean towards having most shots of individual characters, with maybe an over the shoulder shot (see a 2nd character from behind) to make it feel like they are together
@@TheReelRobot❤❤thank you
Apart from UA-cam where can you upload these AI movies?
Do you mean to get exposure? There are some small networks like aifilms.ai and maianetwork.com that promote AI movies. Lots of people use other socials like Instagram. Vimeo and Tubi are good places for amateur films as well, with Tubi kind of being a home for indie films.
@@TheReelRobot Thanks 😊
Your video is really amazing, but I can't find "newbies-30" can you indicate me where can find it please? Thanks in advance
Thanks! It doesn’t have to be “newbies-30” as long as you have a Midjourney channel highlighted on the left side.
Where you go is Discord, make sure you’re in the Midjourney Discord, and on the left side you should see some “rooms”. Some of those rooms say “newbie” and you just click it.
Let me know if you get stuck!
Thank you so much for your help. I will It's a good idea, I'll definitely try it ✌☺✌@@TheReelRobot
Hi, do you give personal tutorial?
Not yet - too busy at the moment. But if you get stuck, you can message me here and I'll reply. And I have courses (including free) coming soon at www.thereelrobot.com where I'll engage with the students personally
The thing about Discord I don’t like is that others on the server can see you work and I don’t want anyone to steal my idea. Is there another way to sidestep them seeing it or no?
Yes. You can create your own discord server (doesn’t cost anything) and do it from there:
eightify.app/summary/computer-science-and-technology/setting-up-your-own-midjourney-server-easy-fast
@@TheReelRobot Thanks
This is such an informative video #businessvideos #socialobster
Do you think we will see a really consistent character option in a year?
For sure. I posted a video about another tool that does it quite well:
ua-cam.com/video/69yQjRGFDDU/v-deo.html
But I know Midjourney is also in the works on their own feature for consistent characters.
Man, this is so unfair. I cant go into an art mayor because i am do afraid ai will take away all my chances. I'm so sad man.
I mean it when I say I believe: this can be better for creatives.
Not for everyone, and not in every way, but the gain is that artists can get out their ideas at a scale never made possible before. Most employed artists I know have done one small job for employers or contract employers, that was never what they really wanted to do. It doesn't let them create their own body of work from their own mind - they're doing menial tasks to spec.
So while I think we will lose some of the things we love about art, and that part is sad, we will have many artists creating art films without $1m budgets and studio approval. It's a net gain that they get to make their own movie, which we otherwise would never have seen in our lifetimes.
@@TheReelRobot it’s not a good thing smh. Look from our perspectives. AI is taking over everything and it’s harder to tell what’s AI or what’s real. Have you seen Project Nightfall’s videos? Have you seen what AI can do?
@@TheReelRobotThen you’re clueless for this situation, you should understand more why people really hate A.I. There’s a lot of abuses, bad actors can use freely, copyright issues, illegal images are also spreading around with the A.I generated. I’m sorry for telling the truth, I’m tired to let a lot of artists, musicians and etc getting bullied by evil toxic lazy dirtbag guys that can’t identify the hard works. 😐
@@Kur404 I'm sorry it's stressing you out, man. Obviously it's not going to be all good, but I don't agree it's all bad. But I don't want to argue given that it's clear this isn't a fun time for you. Hope things work out well for your art.
The big question we ask ourselves: who was the one who created the taboo that Canva is free?
I mean, they even have a page of their website dedicated to how Free it is www.canva.com/free/.
But if you're saying that once you're inside, you get forced to upgrade a lot to use "premium" graphics and stuff, I agree it's mostly not free at that point
@@TheReelRobot thank you very much brother for the explanation
I have try something on my this channel but can't move charector 😂
I want to make a sheet for my own character not an Ai one is that possible?
As in you've hand-drawn a character, and want to have AI help you get more drawings of that character in different angles? What you could do is upload an image to the Midjourney Discord, and copy that image address, then paste it into a prompt where you describe the character but also describing a different angle in the prompt.
You could play with the "--iw" parameter until you get your character from a different angle, then repeat for more angles and collect all your images together. I dunno if it'd work to do a full character sheet in one prompt, but you could try.
Let me know if you meant something else.
Imagine trying to learn an art form and putting actual effort into something original
Most people I’ve met in this space are traditional artists and creators who have adopted AI tools.
Some are people stuck in corporate jobs they hate, often with kids and little time, who see the opportunity to use their artistic tastes in a way where they can juggle other responsibilities. I guess these are the people you hate.
@@TheReelRobot I'm simply against the unauthorized use of another artist's work to create something. It violates copyright law and it's unethical. I do not hate anyone doing this, that would be immature, it is just unethical to use someone else's work and pass it off as your own.
@@mello_bread_art I agree to the extent I’ve come to see all the “in the style of X artist” AI prompting as something that needs to be regulated.
It’s harmless when this stuff isn’t at a quality level where it can actually be monetized easily, but harmful when it gets there.
But otherwise, if it’s blending artistic styles, then it’s no different than you or I being influenced by artists we like and doing something new.
The fact it’s easier than doing it more manually, or being a creative partner, isn’t something we should ever be upset about.
@@TheReelRobot you dont get it, AI observes and takes images from real artists to “learn”, to “train” itself, then it mashes it all up in some generated picture. It’s using those pictures without the artists’ permission. That’s literally copyright violation, which is illegal.