I used a really similar technique to make a little loop a couple of years ago, now when someone asks me how to make something like this i can point them to your tutorial. It explains everything really well
Oh Hi RoseRedTiger! I love your 3D character tutorials and your speedpaints! I know you haven't posted in a while, but I still enjoy watching them! I even recommend them to my friends! You have an interesting process that still helps me too this day. Our lives are busy, but I hope you find time to still paint your gorgeous rose petal Tiger!
True there are few~many techniques, I'm lazy & have to do it on many assets so I use normal map generator like azagaya - Laigter or boundingboxsoftware Materialize
I really wanted to so do animation using live2d or spine pro but now knowing it is possible on blender I don't need those software anymore. Live long Ton Roosendaal and our community.
It kinda depends, on what you actually want to use it for, but yeah its the same idea as what live2d does only you dont need to learn an entirely new program for it.
@@NoraIconiq I even saw some addons made specifically for cut-out animation, some complex rigging systems. Basically mimicking what AE or Live2d can do
@@PolyPaint yeah if its just animation, this works pretty well, the problem is the utility like example one of the popular uses for live2d is making vtuber models. And theres other specialized programs that specifically use files from live2d to work. So if you want to make a vtuber model you'd probably be better off using live2d as it would be easier than trying to convert files. Which from what I heard isnt currently possible. Or atleast a solution hasn't been found. Im still looking into it which is how I found your video.
I've seen many tutorials for animating 2D images in blender using flat images but this is truly next level! I think you're the first to showcase this amazing method, best method ever thank you so much!!
I feel like you have hit jackpot with this tutorial If you ever decide to make a 2.5D in-depth / beginner gumroad tutorial, it is likely to be well recieved, even when priced. Thank you for sharing this!!
Thanks for the support! I think I will prefer spending some time designing such things to offer as much value as I potentially could and also doing some real-life use cases. That might take some time tho
Nice video! It's so interesting to see this technique. I have seen videos from others about creating a 3D model from actually modeling out a 2D character plane (from sphere), to the Grease Pencil, and I'm sure there are lots of other methods as well, something for everyone! I'm just getting into Blender for the first time so it's been very cool to see all the different ways you can create things, it reminds me of back in the early 2010's when I used Photoshop CS4 to animate some GIFs for people's signatures on a forum. Also, 3D programs have come a long way since I used Maya back in 2005, LOL! Thank you so much for the useful tutorial, subbed!
Wow thanks! I am glad to hear that! I use Blender for 4-5 years already and I still discover new things of doing the same task. I personally enjoy more challenges and thinking in a non-conventional ways sometimes. And yes, you can definitely do the same thing in many ways. That's the beauty of freedom and artistic decisions
That’s awesome! I haven’t tried live2d but I feel that the core principle is the same. What could be different is a lack of some specific tools that speed up the process in blender. But I actually saw some rigging addon a for cut-out animation for blender. Maybe that would help
@@PolyPaint it'd be helpful! especially for blender beginners like me (: or having the overlay that shows what keys you're pressing while you record your process.
@@miqueltoast Im using industry standard shortcuts as a previous 3ds max user. So idk how useful that would be for the rest of the world who's using blender native shortcuts XD
Я сто лет пытался найти что-то такое, да все варианты как-то под творческие задумки не ложилось. А вот с блендером это будет значительно проще))) Благо его как раз сейчас начал осваивать. Спасибо огромное, что делитесь своим опытом!!! Побежал смотреть остальной канал :D
You can try this to optimize your workflow: 1. Name your layers with numbering, such as 1_image, 2_image, etc. This ensures the images are ordered correctly. 2. Import all images as planes, setting the Z-depth to be camera-ordered based on alphabetical naming. This allows you to import all your layers in the correct order at once, instead of handling separate layers individually. Essentially, any steps that bring your workflow closer to COA or Spine will help you work faster and more efficiently.
Greate tutorial! Thank you so much!!! I had been wondering for a couple of days now how i could animate a future drawing withoud AF. You couldn’t have posted this at a better time!!! Thank you so much 😊
Most of this can also be done in after effects btw with a depth mask, tho I think making an actual 3d mesh to rig and create depth gives cleaner results.
Nice. Did the same last month for a video. First tried to animate that stuff in Photoshop but it messed up the quality of the elements. Then tried to animate it in Davinci Resolve. And then I remembered I'm actually a 3D artist and no 2D artist. I know how to model, rig and animate 😅 I managed to get stuff done I couldn't have imagined in just only 2D
Hey! I’m just using knife tool(you have it in the left tool panel) in edit mode to cut the plane and add some topology to it. In a way so I can push those points in and out later to create volume.
Very interesting concept. I use depth maps in different ways but never tried animating 2D perspective your way. Such a fresh and cool idea but i think I wouldn't be using it probably- but still great technique, might find use in a future case :))
Hey! I'm just using lasso tools and copy things into new layers, while painting to fill the holes manually. e.g. if you have to cut the character from the background, you will need to fill the hole in the background. The goal is to cover enough so you dont have any holes while its animated. Most of the time, I'm painting using layers, so it's not too hard to separate everything later in the process.
@PolyPaint thank you for the response! I have one more question, after this process , how can i transfer it to Blender?i didnt exactly get the process of the transfer
@@SayoMayo-p2w You can import layers as planes using Add->Image->Mesh Plane (in 4.3) or addon Images as Planes for versions below 4.1. You have to save them previously as png files. So it will be a plane with image texture of your layer. For each layer you'd need a separate plane. In material settings you'd need to enable transparency.
@PolyPaint thank you so so much. You are an incredibly helpful and talented person. I wish you the best and the most success on your youtube career and cant wait to see you blow up :)))
I saw many terms like Motion Graphics, Parallax, Live wallpapers etc. But in the end, its is the same technique used for full-blown 3d animation. the difference is just the shading, animation and rigging is simpler. And the same way as with 3d animation, you could use Face rigs too. The model should be more carefully prepped. you could probably use shapekeys for lips only and then somehow connect that to facial mocap via Iphone. But this is something i have never done. yet
Hi, thx for the tutorial. I don't get your trick to automatically move the points toward the camera (02:07) when in edit mode. In object, it worked because I did set the origin of the object to 3D cursor. But in edit mode how do you set your mesh to use the 3D cursor as an origin of you scalling ?
Ok I've found the solution : make sure to set "Transform pivot point" to "3D cursor". It's the small menu on top of the viewport, close to the proportional editing button.
great work!!! can i ask u for explain? how i can duplicate and adding new png images using node like what u did in 1:15 because when i add new image the first one Disappear ):
@@PolyPaint Hi Poly thank you for you effort. I have unfortunately the same problem as @leloo7. As soon as I add the second alpha image or make a copy from the first and replace the material with the new one, it gets black. Only the "last added alpha image" survives. I used 4.2.3 and 4.3.2. Both the same result 😞. Ok, just found the solution. Both layers of body and head where on the exact same position. If you move one plane just a little bit, it solves my issue. Hoipe this helps others too!
I think MOHO could give a similar result, and for once, actually be *_quicker_* than something else. _lol_ What you did here looks amazing, but for all that work, you might have been better off creating/adapting a figure from Character Creator to make full 3D.
That is true. Although, I really enjoy having everything in one package. Throughout my career I got tired of having dozens of apps to do one single task. I cant even count how many hours of my life I've spent just for export, basically putting things from one basket to another and back again... If I would do this for commercial application I would consider making a library in Blender that would speed up the process drastically.
In the end, I still think it's 90+% the artist and 10% the software. Not sure about Moho specifically, though. I do agree that this was almost already full 3d by the end!
@@AwkwardPain Agree. I overdid the modeling part, it totally depends on the end goal most of the time. For some situations you can make it simpler. For others - do a proper 3d. It's all comes down to resources in the end.
@@PolyPaint Not sure if it's overdone - just that it was almost a 3d model by the end. I still liked the effect, overall and the benefits of keeping your drawing and not having to overlay it on a model is probably saving you a giant headache, tbh!
Hi! I've been following your work for a long time with pleasure, and then this! Cool! Do you have plans to make a step-by-step video of each process for creating such an animation? Something with displaying keys, a detailed explanation of setting keys for video, and so on. Or maybe a paid course? That would be really cool! Thank you and good luck to you in everything
hey! I think doing a course is a cool thing in the end. It needs to be designed well tho, which will take time. Regarding step-by-steps, id love to focus on some of the techniques in more details, showing principles and tool more slowly, but i dont want to go on a "press this button" level. For educational purposes, id love to give options and tool, but i dont want to tell you what to do with those tools, to nourish your own creativity and decision-making skills, which are imo are most important nowadays.
If you think about it, this particular scene would benefit from less animation, specifically around the hands and elbows. It's a rare case where movement breaks the realism, since if you think about it she's now massaging a magically floating ponytail! Just a funny observation.
Most of the time i simply use Emission shader. Its just standard unlit material. To add transparency you have to mix it with Transparent BSDF(top node) where Factor is your Alpha texture. Also, by Default blender scene has some Color Management that makes images appear different. I switch the View Transform to Standard in Scene - Color Management section on the right panel. This way i ensure that colors i see in viewport are exactly the same that i have on my illustration.
Great tutorial! I started following the instructions step by step, but I got stuck when trying to add volume to the points. The projection in the camera isn’t working. Am I doing something wrong in my setup? I'm using Blender 4.2.
I haven't tried 4.2 yet, but i dont think they've changed anything from what I've used. If you're talking about moving points, you have to place 3d cursor to your Camera position, and make the Cursor as transform pivot. Then use scale tool to move scale point towards the camera. Basically, you're sliding points parallel to your camera view vector. Please let me know if that was your question?
@@PolyPaint Thanks for the reply!. I forgot to set the cursor as transform pivot. Now it works for planes but still the problem with the points, when I try to move them with the scale tool it still not working for the camera view I dont know why. Maybe I made a bad Knife work?
Before I rant, I found this video very cool and insightful. Never used blender before but I draw a lot and it would be cool to try it out. But this comment section is really tripping about AI. I’m not sure why people keep simplifying AI to “art theft”, when in reality, it’s just a data synthesizer collecting information to creating its own thing without a manual interaction. Sometimes AI has nothing to do with stealing information or artwork. It’s simply just doing the work for us.
@@phantomgamer9264 it’s not stealing info, it’s collecting data that people willingly put on the internet. It’s no different from a human using references and piecing it together to make their own thing. Plus like I said, AI isn’t just for art or papers. It’s an automated system that can be used for anything without manual control. It’s been around since the early 1900’s, specifically for a chess game, so…
aw, thanks! I will probably do some small vids focused on some tiny things that are easy to follow. Not sure about how detailed that would be, Im using non-standard shortcuts and it will be extra challenge for me. Regarding support, I are supporting me by watching. Later we might create a community or something and come up with other ways to support me in exchange for some value. We'll see how it goes
Sometime in 2025, I'm looking at doing VTubing with Blender and making dinosaur animations using public domain paleoart. I was wondering if I could upload a 2D dinosaur image, make a matching 3D body of it, and then stretch the skin art over it? Like applying a magazine cut-out to a clay body or pouring plaster of paris into a mold, either way. The idea is that I can accept live chat audience art and animate it to do desired shenanigans. This might also allow for AI generated stories with image scenes to become animated sequences, like a comic book come to life. As a person of visual disability I only see out of the top half of my left eye, so I have no symmetry. My hand-eye coordination is absolute garbage, so my facial and body drawings look like they're having a stroke or birth deformity! If I let others do the drawing, I can do the puppeteering.
You can definitely use the projection method onto 3d mesh and project whatever you want. I have a video focused specifically on that method (except im projecting my paintings, but you can project anything this way) Check it out and let me know if you have any questions : ua-cam.com/video/IOuBC1OeODc/v-deo.htmlsi=9H94smYLqkAyhtc8
Oh wow. It has so much potential... is it possible to animate long hair using blender physics tab, like they have in Spine 2d recent update? Ohh, it would be awsome. Right now this method sounds even more intuitive than Spine at least right up untill rigging part (never do rigging😅, but I should try this method!)
I’ve never tried applying physics yet. But I think you could apply it to bones easily. Therefore you could use that simulation with rigged 2.5D model of a hair. Similarly how it’s done in many games. In general, you can apply all the principles that could be applied to full blown 3d animation. The difference is that you have 2.5D model instead of 3d. I’d love to try out some things and make demos for the channel. I think that might be epic
@@PolyPaint oh, thank you so much for the answer! Yeas, totally, it should be fantastic, I wanted to do it for my art for a long time by now and the only thing stopping me so far was the price of Spine. You offer a wonderful alternative
hey! i have a question does images as planes appear on version 4.2? because its not appearing for me and i already checked the add-ons menu too :( the project looks amazing by the way!
Hey! Thanks for watching. I went to check 4.2 version and found out that Image as Planes addon was replaced with native feature. Now you can add images via "Add -> Image -> Mesh" in the top panel when in Object mode
@@PolyPaint btw im really liking your approach. Loving how blender is doing the job of after effects and live2d by itself. With few more tooling, can turn out to be the solution of motion graphics.
Do you think a complete beginner in blender can follow this tutorial. Or first take some course of blender? I jist tried a tutorial of blender and i got lost in shorcuts
Haven’t tried live2d. Principles are the same everywhere, blender is a 3d software so it can do all the stuff that 3d software can do. Not only limited to 2d motion graphics. You can use this method to rig anything the way you want. Mix 3d with 2d, add some fancy shading etc. And it’s free
@@PolyPaintyou basically summed up the advantages I’d say Live2D has. But all the same your tutorial is very useful. As Live2D cubism is 90 Dollars a year and Blender is FREEEEEE! 😀
that is true. I think blender is more flexible in the end as you can mix this type of animation with 3d/2d and in general is much more versatile. But it probably doesn't have most efficient tools for this specific application.
Depends on the task. I made it look pretty complex but it can be applied to simple layers without complex modelling. For example some flat shaded cartoons or puppet-style animation used for games or ads. Also fitting the model to make it look exactly the same will be challenging.
@@PolyPaint I see, yeah that makes sense but could you also please explain how they used this method in the dark alley scene in arcane where everyone moved ? I don't really see it, thank you once again !
@@Unveriefable in those series of shots in Arcane, they used illustrations and just slightly moved them with parallax and maybe subtle animation. The following shot with the car, is 3d already. But this technique is widely used for Backgrounds and its animation. Most of the backgrounds in arcane are 2d, sliced in the same way I showed in a video. some of the parts you can just paint and deform a bit to make it feel animated. For the characters, I’d consider this type of animation as a low budget and quick in production. It has its niche in the industry
This tutorial was great but I think encouraging the use of AI gen was in pretty poor taste. I'll just be hand painting the depth map on my own as I can kinda tell that deep and dark blue = far, and deep dark red = near and green/yellow as the midway point. Thank you for taking the time to make this tutorial.
@PolyPaint i just started blender, this feels like it'll take me longer than a few hours xD. Once you showed all the points and fixing things, I knew it wasn't ez 🤣
@@friendlyneighborhoodartist That is true. This software has pretty steep learning curve. Take small steps and the knowledge will accumulate overtime. At least this is how I approach it, I learn some small bits every day and then I have then at my disposal when I'm working on my projects. Dont be discouraged and keep moving forward! Your toolset will grow with practice and patience.
In every single medium I have ever seen this in, the animation portion of this style always looked tacky to me, it just feels weird and floaty and awkward and lacks proportional continuity, even if hard work was done it just gives me a lazy impression and I don't know why. I have seen it in some productions where it's used stylistically but even then it feels almost worse than even one or two more hand drawn frames, nothing wrong with it at all though as art is art but it's not my style. The frog looked infinitely better to me.
I think it feels cheap because it is in fact cheap. Some companies do illustration for marketing and then integrate then with animation somewhere as afterthought. Reusability in this case is much higher than custom 2d animation. And also, 2d animation and illustration is two completely different beasts. People in illustration can't animate, and people in animation can't render like pro illustrators often. From HR perspective it will be mega difficult to find person to make it look like high-quality illustration and animation simultaneously. Forget about any scalability for such things. Your point is completely valid buy imo it will be quite expensive in the end to make it right and therefore companies go for easier and cheaper solution.
@@PolyPaint This makes a lot of sense to me, in fact I have seen it on a large commercial scale or by triple A companies (and mobile games). For example Hearthstone does it for many of the legendary animated card effects and it works, on a fast large scale and reflects as much but it certainly works for that purpose. Quickly giving illustrations a bit more life.
Sorry to hear that. I tried to give options both for manual and automated approaches. Some depth generators were around for decades, now calling everything Ai is a trend among developers 😅
You have to understand that some AI has been around for decades and is not generative. Aka, it is not trained on data sets stolen from other artists. The is especially important to remember in 3D art because it is denoised using AI and has been since the infancy of AI. The dataset it uses, however, is not stolen from other art but rather it takes the information from the art you’ve made yourself and extrapolates on that to fill in gaps. A good way of explaining it is the empty pixels in a 3D render look around at the other pixels surrounding it to determine what colour it should be. No art theft required. Depth maps, like wise, use the logic that things closest to the camera appear darker than those further away, so it compares the darkest dark of each section to eachother to create a volume, sometimes imperfectly but again, there is not stolen art in the data set. It’s more like using a calculator. These methods have been used in 3D art ever since they were invented to make the Original Toy Story movie by Pixar Animation. The reason Generative AI is bad is because it uses the technology to plagiarize work using stolen unlicensed art without the artists permission. Assuming these old AI methods are bad in this context, is like saying a professional locksmith is evil because when they helped a little old lady with a lost key get back into her house, they used the same lockpick as someone who robbed a house. These are not equivalent uses of the same tech.
@@katelynmurphy7333I’m glad I could help. I’m a painter who’s staunchly against Generative AI but I have started learning to make backgrounds and references in 3D and had to do a ton of research to confront my biases about AI in general as I learned more and more about how 3D art actually works. I’m glad I was able to explain it in a way that makes sense.
When I was in uni.. many years ago, I remember training our own Ai on some input to perform a specific task. That was an exercise in programming a decade ago and had nothing to do with stealing art. Ai now had bad rep specifically because those stolen datasets unfortunately and how aggressively it is pushed onto customers everywhere. Even a washing machine has Ai now for some reason 😅
AI... nah disgusting. Adding your own or someone else's work here... in this way we feed their system to other programs. Especially now using AI is just too risky yk and the speed of work does not appeal to me tho... But Still good tutorial and work.
Yeah, it’s already happening, although in case of manual labour artist decides where to put limits. Less limiting options require more skills and time. But knowing your options to solve the problem is already a big step towards the goals.
I used a really similar technique to make a little loop a couple of years ago, now when someone asks me how to make something like this i can point them to your tutorial. It explains everything really well
Very cool! Thanks!
Oh Hi RoseRedTiger! I love your 3D character tutorials and your speedpaints! I know you haven't posted in a while, but I still enjoy watching them! I even recommend them to my friends! You have an interesting process that still helps me too this day. Our lives are busy, but I hope you find time to still paint your gorgeous rose petal Tiger!
@@notnA51 aww thanks
True there are few~many techniques, I'm lazy & have to do it on many assets so I use normal map generator like azagaya - Laigter or boundingboxsoftware Materialize
Viewers If you use 4.2 version of blender. 11:04 Now bloom effects makes in composing tab with glare modifier. You can find tutorials on youtube
That's a good tip. Im still on 4.1
I really wanted to so do animation using live2d or spine pro but now knowing it is possible on blender I don't need those software anymore.
Live long Ton Roosendaal and our community.
That is so true! Although, specialized software may have some extra tools specifically designed for this type of work
wait... steve?
It kinda depends, on what you actually want to use it for, but yeah its the same idea as what live2d does only you dont need to learn an entirely new program for it.
@@NoraIconiq I even saw some addons made specifically for cut-out animation, some complex rigging systems. Basically mimicking what AE or Live2d can do
@@PolyPaint yeah if its just animation, this works pretty well, the problem is the utility like example one of the popular uses for live2d is making vtuber models. And theres other specialized programs that specifically use files from live2d to work. So if you want to make a vtuber model you'd probably be better off using live2d as it would be easier than trying to convert files. Which from what I heard isnt currently possible. Or atleast a solution hasn't been found. Im still looking into it which is how I found your video.
I've seen many tutorials for animating 2D images in blender using flat images but this is truly next level! I think you're the first to showcase this amazing method, best method ever thank you so much!!
Wow, thanks! Glad it was helpful!
I feel like you have hit jackpot with this tutorial
If you ever decide to make a 2.5D in-depth / beginner gumroad tutorial, it is likely to be well recieved, even when priced.
Thank you for sharing this!!
Thanks for the support! I think I will prefer spending some time designing such things to offer as much value as I potentially could and also doing some real-life use cases. That might take some time tho
Can't believe you can do like this on Blender too! Blender is really getting powerful and what's shocking is, it is freakin FREE!
Exactly. And it’s getting even better.
Also you have software like krita which is also very powerful and free. Makes a great combo
The comment section is kinda weird on this video... the techniques you teach with your videos are amazing! Please keep making blender videos
Thanks a lot! I'm definitely doing more videos
Nice video! It's so interesting to see this technique. I have seen videos from others about creating a 3D model from actually modeling out a 2D character plane (from sphere), to the Grease Pencil, and I'm sure there are lots of other methods as well, something for everyone! I'm just getting into Blender for the first time so it's been very cool to see all the different ways you can create things, it reminds me of back in the early 2010's when I used Photoshop CS4 to animate some GIFs for people's signatures on a forum. Also, 3D programs have come a long way since I used Maya back in 2005, LOL! Thank you so much for the useful tutorial, subbed!
Wow thanks! I am glad to hear that! I use Blender for 4-5 years already and I still discover new things of doing the same task. I personally enjoy more challenges and thinking in a non-conventional ways sometimes. And yes, you can definitely do the same thing in many ways. That's the beauty of freedom and artistic decisions
Thanks for making this...I'm absolutely mindblown. I subscribed
Awesome, thank you!
Thank you, Im learning blender and I wanted to use it as a substitute for Live2D, since it has way more potential. This taught me a lot already.
That’s awesome! I haven’t tried live2d but I feel that the core principle is the same. What could be different is a lack of some specific tools that speed up the process in blender. But I actually saw some rigging addon a for cut-out animation for blender. Maybe that would help
@PolyPaint that's really nice, do you remember the addon name by any chance?
@@wolfgangd3653 I think it was Smart Bones addon. I haven't used it tho. github.com/sketchy-squirrel/smart-bones
blender3D the best point barre 👌🏼
Agreed!
for anyone wondering what the "magic button" is at 4:56, it's Ctrl + P.
Thanks! I wasn't sure if I should include shortcuts, because everyone is using something different usually.
@@PolyPaint it'd be helpful! especially for blender beginners like me (: or having the overlay that shows what keys you're pressing while you record your process.
@@miqueltoast Im using industry standard shortcuts as a previous 3ds max user. So idk how useful that would be for the rest of the world who's using blender native shortcuts XD
@@PolyPaint OHHH ok yeah that makes sense...
That's so cool! I've been using blender for years, but never thought of trying this kind of effect on it. Looks awesome
Thanks! Yeah it’s really cool thing to do, and can be applied literally to anything, character, prop, environments
I had been thinking about this posdibility for awhile, glad someone did it, and confirmed how I thought it would work. Nicely done.
Great! I'm glad it helps
Finally a new video :) Very interesting and informative! Thank you
Thank you brother!
Я сто лет пытался найти что-то такое, да все варианты как-то под творческие задумки не ложилось. А вот с блендером это будет значительно проще)))
Благо его как раз сейчас начал осваивать.
Спасибо огромное, что делитесь своим опытом!!! Побежал смотреть остальной канал :D
Класс. Спасибо!
moreeee 2d plus 3d tuts pleaseeee loving it
More to come!
Your art style is brilliant 💕😘
Thank you so much 😀
You can try this to optimize your workflow:
1. Name your layers with numbering, such as 1_image, 2_image, etc. This ensures the images are ordered correctly.
2. Import all images as planes, setting the Z-depth to be camera-ordered based on alphabetical naming.
This allows you to import all your layers in the correct order at once, instead of handling separate layers individually.
Essentially, any steps that bring your workflow closer to COA or Spine will help you work faster and more efficiently.
Awesome tips! Thanks!
Greate tutorial! Thank you so much!!! I had been wondering for a couple of days now how i could animate a future drawing withoud AF. You couldn’t have posted this at a better time!!! Thank you so much 😊
Glad I could help! This is where I took my inspiration from. I was always fascinated by those animated splash arts from League of Legends.
bakan hermano, super buen tutorial
Thanks!
wow great video! now i just need to get good at painting and get creative
Thanks! Yeah, that's the skillset that will greatly benefit you and your art. Worth learning imo
Most of this can also be done in after effects btw with a depth mask, tho I think making an actual 3d mesh to rig and create depth gives cleaner results.
True! Initially this style was dominated by AE. Now we have lots of different options
If you make a cone or any shape and in the material you connect an emission shader to the volume you can make god rays quite easily.
If you see that it looks low res you can increase the volume resolution in the render settings
cool tip, i will try that out!
This tutorial is brilliant and exactly what I needed right now! Thanks, instant subscribe 😊
Awesome, thank you!
Beautiful tutorial- the butterflies are animated in reverse though!
Thank! Yeah you’re right, people brought that to my attention, I haven’t even noticed that myself 😅
Ground Breaking, Thnx so much this has opened so many doors
Glad it was helpful!
Nice. Did the same last month for a video.
First tried to animate that stuff in Photoshop but it messed up the quality of the elements.
Then tried to animate it in Davinci Resolve.
And then I remembered I'm actually a 3D artist and no 2D artist. I know how to model, rig and animate 😅 I managed to get stuff done I couldn't have imagined in just only 2D
Great story! In the end it got resolved and it’s all that matters! It’s really empowering when you’re not limited to one medium.
i was not expecting the whisper at the end saying "subscribe"
Haha. I hope it worked XD
Great video! I would love to see you rig and animate a full body 2D character in Blender.
Thanks! That would be cool
woah never knew this was possible on blender, tysm for the tutorial!
Happy to help!
What exactly are you doing from 1:19 on, how does the image splitting work?
Hey! I’m just using knife tool(you have it in the left tool panel) in edit mode to cut the plane and add some topology to it. In a way so I can push those points in and out later to create volume.
Awesome tutorial! Can't wait for your channel to blow up 👍👍
Thanks!
Awesome content Sergey! Cant wait to try :) Spasibo!
Haha! My pleasure!
Very interesting concept. I use depth maps in different ways but never tried animating 2D perspective your way. Such a fresh and cool idea but i think I wouldn't be using it probably- but still great technique, might find use in a future case :))
Thank you I am so glad that I come across this video and your channel
I'm so glad you’re here too!
Really cool techniques here and definitely something I will try out in Blender, thanks for the inspiration! :)
My pleasure!
This is a very cool tutorial😯 and thanks for the source. I really want to repeat it😊
Glad you like it!
YES, I finally don't have to pay for spine 2D no more 🙏🙏🙏
Awesome! Blender is worth learning! Also, check blender market for 2d cut-out animation. I think I saw some addons for that. 💪
Wow, this is what I needed 👏
Awesome!
Hey, could you explain how you seperate/cut the image into several pieces in krita?
@1:01
Hey! I'm just using lasso tools and copy things into new layers, while painting to fill the holes manually.
e.g. if you have to cut the character from the background, you will need to fill the hole in the background. The goal is to cover enough so you dont have any holes while its animated.
Most of the time, I'm painting using layers, so it's not too hard to separate everything later in the process.
@PolyPaint thank you for the response!
I have one more question, after this process , how can i transfer it to Blender?i didnt exactly get the process of the transfer
@@SayoMayo-p2w You can import layers as planes using Add->Image->Mesh Plane (in 4.3) or addon Images as Planes for versions below 4.1. You have to save them previously as png files.
So it will be a plane with image texture of your layer.
For each layer you'd need a separate plane. In material settings you'd need to enable transparency.
@PolyPaint thank you so so much. You are an incredibly helpful and talented person. I wish you the best and the most success on your youtube career and cant wait to see you blow up :)))
Thank you! This was a bit of a confusing part in the video so i appreciate the explanation@@PolyPaint
I work with this kind of animation and your tutorial its great! Keep doing with the great work!
Thanks!
que hermoso tutorial, eres un genio.
Thank you!
HOLY SHIT! well done, im subscribing!
Thanks for watching and for subscription!
Love it. Very nice for scenes with less movement. Does this technique have a name ? Can you LipSynch onto it ? *Subscribed*
I saw many terms like Motion Graphics, Parallax, Live wallpapers etc. But in the end, its is the same technique used for full-blown 3d animation. the difference is just the shading, animation and rigging is simpler. And the same way as with 3d animation, you could use Face rigs too. The model should be more carefully prepped. you could probably use shapekeys for lips only and then somehow connect that to facial mocap via Iphone. But this is something i have never done. yet
Great tutorial. I will try this. Backward butterflies 😸
Thank you! Cheers!
That was truly great! 👍🏼
Glad you liked it!
awesome! Thanks!
My pleasure!
Hi, thx for the tutorial. I don't get your trick to automatically move the points toward the camera (02:07) when in edit mode. In object, it worked because I did set the origin of the object to 3D cursor. But in edit mode how do you set your mesh to use the 3D cursor as an origin of you scalling ?
Ok I've found the solution : make sure to set "Transform pivot point" to "3D cursor". It's the small menu on top of the viewport, close to the proportional editing button.
@@jlnjnn4279 Yes, exactly. I think i had that in the video although I will do a better job highlighting these things in my future videos
Holly shit, that was I needed for my personal project, thank you very much.
Glad to hear that!
i could defintely use this techique in my work.
Awesome!
great work!!! can i ask u for explain? how i can duplicate and adding new png images using node like what u did in 1:15 because when i add new image the first one Disappear ):
Thanks! I create a copy of material for each layer of illustration and change the image texture for each material! I hope that helps!
@@PolyPaint i will try thank u so much
@@PolyPaint Hi Poly thank you for you effort. I have unfortunately the same problem as @leloo7. As soon as I add the second alpha image or make a copy from the first and replace the material with the new one, it gets black. Only the "last added alpha image" survives. I used 4.2.3 and 4.3.2. Both the same result 😞.
Ok, just found the solution. Both layers of body and head where on the exact same position. If you move one plane just a little bit, it solves my issue. Hoipe this helps others too!
@@AIDreamWriter Awesome! Yeah, I'm adding tiny offset too so Blender isn't confused what to render first.
damn now i can make blue lock season 2
lmao XD
Im Dying 🤣🤣🤣
I think MOHO could give a similar result, and for once, actually be *_quicker_* than something else. _lol_ What you did here looks amazing, but for all that work, you might have been better off creating/adapting a figure from Character Creator to make full 3D.
That is true. Although, I really enjoy having everything in one package. Throughout my career I got tired of having dozens of apps to do one single task. I cant even count how many hours of my life I've spent just for export, basically putting things from one basket to another and back again... If I would do this for commercial application I would consider making a library in Blender that would speed up the process drastically.
In the end, I still think it's 90+% the artist and 10% the software. Not sure about Moho specifically, though. I do agree that this was almost already full 3d by the end!
@@AwkwardPain Agree. I overdid the modeling part, it totally depends on the end goal most of the time. For some situations you can make it simpler. For others - do a proper 3d. It's all comes down to resources in the end.
@@PolyPaint Not sure if it's overdone - just that it was almost a 3d model by the end. I still liked the effect, overall and the benefits of keeping your drawing and not having to overlay it on a model is probably saving you a giant headache, tbh!
Hi! I've been following your work for a long time with pleasure, and then this! Cool! Do you have plans to make a step-by-step video of each process for creating such an animation? Something with displaying keys, a detailed explanation of setting keys for video, and so on. Or maybe a paid course? That would be really cool! Thank you and good luck to you in everything
hey! I think doing a course is a cool thing in the end. It needs to be designed well tho, which will take time. Regarding step-by-steps, id love to focus on some of the techniques in more details, showing principles and tool more slowly, but i dont want to go on a "press this button" level. For educational purposes, id love to give options and tool, but i dont want to tell you what to do with those tools, to nourish your own creativity and decision-making skills, which are imo are most important nowadays.
If you think about it, this particular scene would benefit from less animation, specifically around the hands and elbows. It's a rare case where movement breaks the realism, since if you think about it she's now massaging a magically floating ponytail! Just a funny observation.
Couldn’t agree more. It’s a process of learning that never ends. Next one I do should be better 🔥
Insane!
Thanks!
great video as always
I appreciate that!
that was a nice video! thank you
Thank you too!
mad impressive, exactly what i was looking forward to. thank you so much!
Glad I could help!
ah another masterpiece
yur butterflies are reverse flyers!
😆
I made this with my artwork too. Hello fellow blender user👋
10:34 Dude just made an ai video WITHOUT ai. Mad respect
lol. This is the way!
Cool
What is the material settings for your image? How is it so vibrant and true to original illustration?
Most of the time i simply use Emission shader. Its just standard unlit material. To add transparency you have to mix it with Transparent BSDF(top node) where Factor is your Alpha texture.
Also, by Default blender scene has some Color Management that makes images appear different. I switch the View Transform to Standard in Scene - Color Management section on the right panel.
This way i ensure that colors i see in viewport are exactly the same that i have on my illustration.
Thanks!
Welcome!
Great tutorial! I started following the instructions step by step, but I got stuck when trying to add volume to the points. The projection in the camera isn’t working. Am I doing something wrong in my setup? I'm using Blender 4.2.
I haven't tried 4.2 yet, but i dont think they've changed anything from what I've used. If you're talking about moving points, you have to place 3d cursor to your Camera position, and make the Cursor as transform pivot. Then use scale tool to move scale point towards the camera. Basically, you're sliding points parallel to your camera view vector.
Please let me know if that was your question?
@@PolyPaint Thanks for the reply!. I forgot to set the cursor as transform pivot. Now it works for planes but still the problem with the points, when I try to move them with the scale tool it still not working for the camera view I dont know why. Maybe I made a bad Knife work?
Topology plays a big role yeah. It has to be somewhat classical 3d topology for best results.
Before I rant, I found this video very cool and insightful. Never used blender before but I draw a lot and it would be cool to try it out. But this comment section is really tripping about AI. I’m not sure why people keep simplifying AI to “art theft”, when in reality, it’s just a data synthesizer collecting information to creating its own thing without a manual interaction. Sometimes AI has nothing to do with stealing information or artwork. It’s simply just doing the work for us.
Where do you think it gets it's info from? Stealing
Glad that my video is inspiring! Try it out! I really love blender ☺️
@@phantomgamer9264 it’s not stealing info, it’s collecting data that people willingly put on the internet. It’s no different from a human using references and piecing it together to make their own thing. Plus like I said, AI isn’t just for art or papers. It’s an automated system that can be used for anything without manual control. It’s been around since the early 1900’s, specifically for a chess game, so…
i love you , i just started and was searching for something like this for so long
aw, thanks! I will probably do some small vids focused on some tiny things that are easy to follow. Not sure about how detailed that would be, Im using non-standard shortcuts and it will be extra challenge for me. Regarding support, I are supporting me by watching. Later we might create a community or something and come up with other ways to support me in exchange for some value. We'll see how it goes
Sometime in 2025, I'm looking at doing VTubing with Blender and making dinosaur animations using public domain paleoart.
I was wondering if I could upload a 2D dinosaur image, make a matching 3D body of it, and then stretch the skin art over it? Like applying a magazine cut-out to a clay body or pouring plaster of paris into a mold, either way.
The idea is that I can accept live chat audience art and animate it to do desired shenanigans. This might also allow for AI generated stories with image scenes to become animated sequences, like a comic book come to life.
As a person of visual disability I only see out of the top half of my left eye, so I have no symmetry. My hand-eye coordination is absolute garbage, so my facial and body drawings look like they're having a stroke or birth deformity! If I let others do the drawing, I can do the puppeteering.
You can definitely use the projection method onto 3d mesh and project whatever you want. I have a video focused specifically on that method (except im projecting my paintings, but you can project anything this way) Check it out and let me know if you have any questions :
ua-cam.com/video/IOuBC1OeODc/v-deo.htmlsi=9H94smYLqkAyhtc8
Nice video. Subscribed
Thanks for the sub!
Oh wow. It has so much potential... is it possible to animate long hair using blender physics tab, like they have in Spine 2d recent update?
Ohh, it would be awsome.
Right now this method sounds even more intuitive than Spine at least right up untill rigging part (never do rigging😅, but I should try this method!)
I’ve never tried applying physics yet. But I think you could apply it to bones easily. Therefore you could use that simulation with rigged 2.5D model of a hair. Similarly how it’s done in many games.
In general, you can apply all the principles that could be applied to full blown 3d animation. The difference is that you have 2.5D model instead of 3d.
I’d love to try out some things and make demos for the channel. I think that might be epic
@@PolyPaint oh, thank you so much for the answer! Yeas, totally, it should be fantastic, I wanted to do it for my art for a long time by now and the only thing stopping me so far was the price of Spine. You offer a wonderful alternative
👍!!~~~~NICE.
am gonna try this, maybe i can move on from spine pro
GOAT
🐐
Дякую:3
Нема за що!
hey! i have a question does images as planes appear on version 4.2? because its not appearing for me and i already checked the add-ons menu too :(
the project looks amazing by the way!
Hey! Thanks for watching. I went to check 4.2 version and found out that Image as Planes addon was replaced with native feature. Now you can add images via "Add -> Image -> Mesh" in the top panel when in Object mode
@PolyPaint thank you so much!
Hello there @PolyPaint, I downloaded the blend file, but it does not open.
Hello. What blender version are you using? It was created in blender 4.1. Older versions of blender won’t open it
@ I recently upgraded to 4.2 I tried it there but the file opens then the system closes before even loading.
@@ThomasAlanWayne wow. Im totally guessing rn, but this might be some hardware issue? lack of gpu or ram? idk honestly.
how do you setup lightning?
I dont have any lighting here. It was painted and im using emission (unlit) shader. The only thing blender adds is Bloom effect
@@PolyPaint btw im really liking your approach. Loving how blender is doing the job of after effects and live2d by itself. With few more tooling, can turn out to be the solution of motion graphics.
Thanks! Yeah, I love blender for that. It’s super versatile and can do almost anything. And it’s getting better every update
I need a noob tutorial plssssss😭😭
I already have some tutorials on my channel. There isn’t much more that I can show. In 2d I use the same techniques over and over again
are the butterflies flying backwards?
lol yeah. Isn’t that the way they usually fly? 🤪
@@PolyPaint 😆the longbig part of the wing is suppose to be at the top when it's flying
@@NoZeroGG haha! thanks. I messed up that part. didnt even notice XD
Do you think a complete beginner in blender can follow this tutorial. Or first take some course of blender? I jist tried a tutorial of blender and i got lost in shorcuts
I don’t think you’ll be able to follow if you are complete beginner. At least you need some familiarity with the software and some art background
difference between this and live2d?
Haven’t tried live2d. Principles are the same everywhere, blender is a 3d software so it can do all the stuff that 3d software can do. Not only limited to 2d motion graphics. You can use this method to rig anything the way you want. Mix 3d with 2d, add some fancy shading etc. And it’s free
Im a blender beginner and would appreciate if there was a much more in depth tutorial on this... im a 2d artist lol
Yeah, i think soma basics tutorial are getting higher in my priorities. Thanks for the feedback!
literally live2dcubism but in blender
Haven't tried it, but looks like it's using same principles. Might be more streamlined and have more tools for that specific type of art.
@@PolyPaintyou basically summed up the advantages I’d say Live2D has. But all the same your tutorial is very useful. As Live2D cubism is 90 Dollars a year and Blender is FREEEEEE! 😀
3:10 🧿🐽🧿
you saw that XD
it's brilliant but spine2d is dedicated for this kind of stuff
that is true. I think blender is more flexible in the end as you can mix this type of animation with 3d/2d and in general is much more versatile. But it probably doesn't have most efficient tools for this specific application.
a batterfly movin backwards😢
oopsie 😅
Yall are catching up, but still havent figured out my tools kekeke
Hello bro, why not just make a 3D model ? honestly asking, like to me it seems like making a 2D reference 3D with extra steps, thanks for your time!
Depends on the task. I made it look pretty complex but it can be applied to simple layers without complex modelling. For example some flat shaded cartoons or puppet-style animation used for games or ads. Also fitting the model to make it look exactly the same will be challenging.
@@PolyPaint I see, yeah that makes sense but could you also please explain how they used this method in the dark alley scene in arcane where everyone moved ? I don't really see it, thank you once again !
@@Unveriefable in those series of shots in Arcane, they used illustrations and just slightly moved them with parallax and maybe subtle animation. The following shot with the car, is 3d already. But this technique is widely used for Backgrounds and its animation. Most of the backgrounds in arcane are 2d, sliced in the same way I showed in a video. some of the parts you can just paint and deform a bit to make it feel animated. For the characters, I’d consider this type of animation as a low budget and quick in production. It has its niche in the industry
@@PolyPaint Thank you for the explanation, God bless you and I wish you a wonderful day!
Jump scare 3:23
lol. You’ve noticed 🔥
This tutorial was great but I think encouraging the use of AI gen was in pretty poor taste. I'll just be hand painting the depth map on my own as I can kinda tell that deep and dark blue = far, and deep dark red = near and green/yellow as the midway point. Thank you for taking the time to make this tutorial.
If not take automation into account, I would rather skip the depth painting completely and sculpt the shape in blender directly.
Брат, можно таким обращом через ИИ делать мультики?
I don’t know. Haven’t tried, but some ppl do generate videos
Eshe rano, cherez goda dva navernoe
bruh this tutorial basically is like "way to do this for a whole month" lol. not easy at ALL XD
It took me a few hours, to be honest. Not as hard as it might seem.
@PolyPaint i just started blender, this feels like it'll take me longer than a few hours xD. Once you showed all the points and fixing things, I knew it wasn't ez 🤣
@@friendlyneighborhoodartist That is true. This software has pretty steep learning curve.
Take small steps and the knowledge will accumulate overtime. At least this is how I approach it, I learn some small bits every day and then I have then at my disposal when I'm working on my projects.
Dont be discouraged and keep moving forward! Your toolset will grow with practice and patience.
In every single medium I have ever seen this in, the animation portion of this style always looked tacky to me, it just feels weird and floaty and awkward and lacks proportional continuity, even if hard work was done it just gives me a lazy impression and I don't know why. I have seen it in some productions where it's used stylistically but even then it feels almost worse than even one or two more hand drawn frames, nothing wrong with it at all though as art is art but it's not my style. The frog looked infinitely better to me.
I think it feels cheap because it is in fact cheap. Some companies do illustration for marketing and then integrate then with animation somewhere as afterthought. Reusability in this case is much higher than custom 2d animation. And also, 2d animation and illustration is two completely different beasts. People in illustration can't animate, and people in animation can't render like pro illustrators often. From HR perspective it will be mega difficult to find person to make it look like high-quality illustration and animation simultaneously. Forget about any scalability for such things.
Your point is completely valid buy imo it will be quite expensive in the end to make it right and therefore companies go for easier and cheaper solution.
@@PolyPaint This makes a lot of sense to me, in fact I have seen it on a large commercial scale or by triple A companies (and mobile games). For example Hearthstone does it for many of the legendary animated card effects and it works, on a fast large scale and reflects as much but it certainly works for that purpose. Quickly giving illustrations a bit more life.
lost me at AI... =/
Sorry to hear that. I tried to give options both for manual and automated approaches. Some depth generators were around for decades, now calling everything Ai is a trend among developers 😅
You have to understand that some AI has been around for decades and is not generative. Aka, it is not trained on data sets stolen from other artists. The is especially important to remember in 3D art because it is denoised using AI and has been since the infancy of AI. The dataset it uses, however, is not stolen from other art but rather it takes the information from the art you’ve made yourself and extrapolates on that to fill in gaps.
A good way of explaining it is the empty pixels in a 3D render look around at the other pixels surrounding it to determine what colour it should be. No art theft required.
Depth maps, like wise, use the logic that things closest to the camera appear darker than those further away, so it compares the darkest dark of each section to eachother to create a volume, sometimes imperfectly but again, there is not stolen art in the data set.
It’s more like using a calculator.
These methods have been used in 3D art ever since they were invented to make the Original Toy Story movie by Pixar Animation.
The reason Generative AI is bad is because it uses the technology to plagiarize work using stolen unlicensed art without the artists permission.
Assuming these old AI methods are bad in this context, is like saying a professional locksmith is evil because when they helped a little old lady with a lost key get back into her house, they used the same lockpick as someone who robbed a house.
These are not equivalent uses of the same tech.
@@mickymcbryan4814 This is a a very insightful explanation, thank you for taking the time! :)
@@katelynmurphy7333I’m glad I could help. I’m a painter who’s staunchly against Generative AI but I have started learning to make backgrounds and references in 3D and had to do a ton of research to confront my biases about AI in general as I learned more and more about how 3D art actually works. I’m glad I was able to explain it in a way that makes sense.
When I was in uni.. many years ago, I remember training our own Ai on some input to perform a specific task. That was an exercise in programming a decade ago and had nothing to do with stealing art. Ai now had bad rep specifically because those stolen datasets unfortunately and how aggressively it is pushed onto customers everywhere. Even a washing machine has Ai now for some reason 😅
AI... nah disgusting. Adding your own or someone else's work here... in this way we feed their system to other programs. Especially now using AI is just too risky yk and the speed of work does not appeal to me tho... But Still good tutorial and work.
I prefer manual work too tbh, but knowing your options is always good
its very limiting, AI will take over easily
Yeah, it’s already happening, although in case of manual labour artist decides where to put limits. Less limiting options require more skills and time. But knowing your options to solve the problem is already a big step towards the goals.
thanks for the tutorials :D
No problem!