Easily Turn Videos into AI ANIMATION ! | After Effects & Stable Diffusion Control Net
Vložit
- čas přidán 10. 07. 2024
- Get Surfshark VPN at Surfshark.deals/MAXNOVAK and enter promo code MAXNOVAK for 83% off and 3 extra months for FREE!
▶My Digital Asset Store / Presets / Plugins / & More!: mediamonopoly.co
(business inquiries: sawickimx@gmail.com)
NEW Method for INSANE VIDEO to AI ANIMATION !
------------------------------------------------------------------------------------------
►Follow me on IG: @_maxnovak
►Join Our Community Discord: / discord
(Editing Help, Job Offers, Submit Videos, Interact!)
------------------------------------------------------------------------------------------
LINKS:
Start here donwload stable diffusion webui: github.com/AUTOMATIC1111/stab...
Download Python: www.python.org/downloads/
Control Net Models: huggingface.co/lllyasviel/Con...
Download custom models (optional): huggingface.co/models?other=s...
More custom models: civitai.com/
Flowframes: nmkd.itch.io/flowframes
EBsynth: ebsynth.com/
Channels mentioned: • ANIME ROCK, PAPER, SCI...
/ @enigmatic_e
Today I'll show you an easy to use and powerful Ai processing tool that is 100% free to download: Stable Diffusion Img2Img ! I'll show you how to setup and install, how to create video animations using batch processing in SD Img2Img. How to take your Ai animations and combine them with your footage in After effects & More! Enjoy!
TIMESTAMPS:
0:00 - Intro
0:43 - Sponsor
1:31 - Installing Updated Stable Diffusion
3:15 - Preparing video you want to animate with AI
4:15 - Explanation and Installation of Control Net
6:28 - OPTIONAL BONUS Custom Model Sets
8:31 - How to Use Stable Diffusion Img2Img
12:30 - Control Net with Stable Diffusion
15:00 - Batch export back into After Effects
16:30 - Methods for cleaning up jitter
18:45 - Ebsynth method
20:25 - Example of what you can do when combining different styles with Ai
21:13 - Outro
-----------------------------------------------------------------------------------------
#aftereffects #stablediffusion #tutorial - Krátké a kreslené filmy
There is a ton of work in this tutorial, and it kept me watching for 20 minutes straight. Congrats man
when you constantly run AI on a computer basically 24/7 (which not gonna lie, happens to a lot of people finding out they can auto gen images) it's easy to get stuck in a loop and forgetting the other capabilities
MAX NOVAK 🐐🐐
Best tutorial of Videos to Anime. Love it. Keep going bro.
do you disable to control nets when you batch export? otherwise it just exports like 3 different frames for each one frame
Nice! Thanks for the shout out again.
Great stuff. Keep em coming
Damn this tutorial is fast and right to the point. Thx a lot for this work 👏👏
When using control net the export gives me duplicate images - one with outlines and the rendered image.
wow dude great that little line of code about the xformers makes lunch faster and straight to the interface, thanks✌
Best tutorials online even though I haven't tried yet. Thanks!
You’re amazing. Continue to be that 👍🏼💯
amazing stuff, thanks for making the steps clear
How did he open the Stable Diffusion on browser?
Excellent explanation! Thank you
Thank you Max!
thx ur the master of help for creators
incredible ! good explain thank you
Thanks for covering this. I've been trying to get into AI enhanced video. The LSD trip effect was annoying me haha. Testing this out.
Great video and explanation
This is a game changer. Unbelieveable stuff.
Brilliant!🍺👽
thanks for this great tutorial. But I have a problem that my control net is not working. After I do the same setting and click generate, the map is not coming out. How to fix this?
Cool chit. Keep ‘em coming.
You the man🙏🏼
what i love about Stable Diffusion tutorials is that almost Everyone says that it's free, but almost Nobody mentions that in order to make it not frustrating to use, you need a monster gpu like 3090ti or so. But i love it in a saddistic way.
lmao facts it fried my computer trying to use that
cap max is using a rtx 2080
I have used a 3060 and it does well, but ive decided against using stable diffusion too much, i dont want my graphics card to age too qickly
@@moosterbeast it works the same as playing game son your pc there’s not much difference
@@DrippyVisuals Really? Maybe when its batch processing, I have nothing to do so I noticed the fan working overtime instead of the games when I'd be busy playing and not notice it. I'll check next time I play TLOU .
Nice tutorial! One point I missed was the recomendation to download the .safetensors instead of the .ckpt files 👌
watching from Grenada in the Caribbean
is says Couldn't launch python even if I already install phyton
please help
One question how do you get the blue interface instead of the white one? thank u!
Thank you so much my dude ❤❤❤
Thank you so much. I learned so much. For people who don't have Adobo Affect Effects, you can use other free software
how do I find and install all those Controlnet Preprocessors, I have a completely diffferent set of names...
When I batch-export, it exports the final image + the control net maps. Is there a way for exporting only the final results?
AAMMMAAAAZZZINNNNGGGG ❤❤❤ 🔥🔥🔥🔥🔥🔥🔥🔥🔥
brother cuando abro el archivo por lotes no me aparece instalando me aparece no python, hise todo al pie del video
hey guys.. i want to add an Image Init to Deforum in A1111 but i get always this error:
Error: 'Init image path or mask image path is not valid'. Check your schedules/ init values please. Also make sure you don't have a backwards slash in any of your PATHs - use / instead of \. Full error message is in your terminal/ cli.
Image path is: C:/webui/images/01.png
i tried it with backslash, upload the image on the web and use a web link, tried another image, another path.. nothing work :(
how do you get that hed option for preprocessor in control net? i downloaded all the models
i think it was renamed to softedge_hed in an update
Welcome to the party, which will be much different 3 weeks from now. Glad you got to stop in for a few minutes, lol.
Holy shit, this is amazing thanks for sharing
mine is rendering out the HED map as well not sure how to seperate for AE
strange, can you screenshot your settings?, maybe you need to select the image not the map, perhaps you have both selected
Hi Max! How to make it take into account the depth map and canny, but not output them to a folder with shared images. In your video, only ready-made images are obtained in the "out" folder, and I also have depth and canny saved there!
Same issue here.
Did you find a solution to this?
@@braedongarner yes, in the "control no" itself, leave a free window, without a picture, and it will always generate new ones.
@@phosphene8120 excellent, thanks for your help!
Thank you!
Well dayum! 👀
guys, for those that have the python issue, make sure to check "path" on the very first step of the python setup and that you install the python that is linked in the stable diffusion download link and not the .11 one that max shows, it's not compatible
gracias dude
Thanks! Important step!!!
Greta video, is there anyway to acheive the same results but using Midjourney instead of stable diffusion?
Hi my controlnet doesnt show any models in the right side panel it shows only processor
When I interrogate the image, after about 2 or 3 mins it just says error
For some reason I can't slide the control net slider to get more.
Manually adding the number doesn't work.
man i have a problem when i generate my video like you on 14:38 it also saves the preview images from controlnet how should i disable that?
me too, anyone got a solution?
@@davewoodhead2452 Did anybody figure out how to get rid of these preview images (canny, depth, etc.)?
@@braedongarner I’m on a Mac and use the the Terminal command rm *-1.png and then rm *-2.png with Terminal located at the relevant directory using the cd command. Take care because they are totally removed and not recoverable from the Trash.
very good tutorial. I got a big picture of SD .. however most link is for windows .. Os can not use...
How much would you charge someone to make these edits? How do you go about pricing these different effects? They can get so sophisticated...
Hey Max, this is a truly mind boggling technology. I've been able to implement stable diffusion to my videos however the consistency is rather lacking if you were to try to create a video about 2-3 minutes long. If either you or someone who has been using this technique could help me I would appreciate it. What is the secret to having consistent motion throughout the frames in the output directory?
my 1650ti is crying
Use the same seed from your satisfied result
i got the same gpu. have you by any chance tried using stable diffusion on ur pc with those specs? did it work?@@Cineinsaner
but can anyone tell me how much VRAM do i need to run this?
been playing with this now for a short time, holy fucking shit dude this is the biggest game changer, thanks so much for covering this!
Couldn't launch python
exit code: 9009
stderr:
Python was not found; run without arguments to install from the Microsoft Store, or disable this shortcut from Settings > Manage App Execution Aliases.
Launch unsuccessful. Exiting.
Press any key to continue . . . how to fix this??
did you fix it?
Why is that your batch rendering came out with only 1 photo. When i did my batch rendering, each frame was rendered 3 times. One was the AI art itself, then the 2nd image was the same frame but the control net HED image and the third render was the same frame but the canny image? I am following to a T but somehow my export was different.
I don't have any model on the right side after I select the preprocessor... why is that?
im facing an error
RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'
how to resolve this? im unable to find the problem in the whole internet
When I try to use control net this pops up: RuntimeError: Placeholder storage has not been allocated on MPS device!
Can anyone assist with this?
i can't run it
ERROR: Could not find a version that satisfies the requirement torch==1.13.1+cu117 (from versions: 2.0.0, 2.0.0+cu117)
ERROR: No matching distribution found for torch==1.13.1+cu117
can anyone help me
I am also facing the same error
same
Uninstall and extract zip file again. Basically start over and use Python version 3.10.6
My output also is saving my lines canny/pose images as well.
i've problems with automate 1111, in my computer these programs are not stable
when I click on that batch file I get this error: " Couldn't launch python exit code: 9009" any solutions???
samee
change the path directory of python
go to your local disc where you've installed your python, generally it's in the C' disc
then enable the show hidden files, under view option
then go to users - click on your named account - appdata - local - program - python - copy the path directory
then search for edit enviroment variables in the window tab - click on path - add - paste the path followed by "\"
then add another directory - paste the same followed by '\Scripts\
move these two directories at the top
click ok
i hope this works.
Small deviations in this Joker test is ok, it is his nature :D
cant we do this in deforum
thanks for the video.. my bat file cannot locate python. I first installed python with paths checked. and then modified the bat as in your video but every time I run it it cannot proceed. cannot find bat file. any solve to this/> sorry for the noobie Q I tried repairing uninstall re install python.. any ideas?
isnt safetensors better than ckpt because of the risk of hackers/viruses?
The output video made me realize that Jeff Goldblum would have made an excellent joker when he was younger
The new De Staat music video is made like that !!
Hello, Python can't be found in my PC with the webi-user.. do you know how to fix it ?
I tried with Python's version from the video and the newest one
same for me
does this work on Mac?
Hello! Thank you very much for your teaching video. I have learned a lot. If you are in the process of making a teaching video, don’t startle the mouse. This will be more conducive to the protagonist of our lecture program.
what lol
how much space it need?
Tq
dope
ur gpu?
Better to wait content from u it’s always complicated tnx
please help me i followed step by step and got error code 1 ERROR: cannot find 'git' - do you have git installed and in your path ?
I have git installed in general but I didn’t clone the repository I downloaded the zip so not sure why there’s a git error
MY WEB UI looks completely sidderent in the read me there is no where to put the code
error could not install torch
Very cool but also a bit more complex than anything I want to do in AI as far taking that much time to adjust settings etc..., Im here for the AI to do the AI and automate things like these. I think we are getting closer. Great video though, just sharing my preference as we glide into the world of AI.
add control net and making a custom lora based on video
Does anyone know the name of the model which is used in the thumbnail?
it's Jimin from bts
OMG! this is complicated to me, im going crazy, i wish i understand and learn everything like everyone, damn :(
No way to do this on Mac?
Hey, What are your PC specs?
💫
Error code: 128
stdout:
stderr: fatal: not a git repository (or any of the parent directories): .git
Why sir?
Batch doesn;'t work and keeps getting errors
Does it have AI, not like those enhanced videos, you know, one more like those like diffusion or lora, that I can put a scene from a movie, then it generates the same video but following the text of the prompt, like a For example, an iron man scene, I want him to wear super men's clothes and so on, it's like it's for photos, but with videos, do you know something, friend?
Control net isn't showing in my stable diffusion
nvm got it. Had to re-launch
Did you skip something at 4:05?
How to open Stable Diffusion on my browser?
--xformers--autolaunch doesn't work anymore me thinks
Is all of this completely free?
Hey man, I have a problem. I followed all the steps for Mac and I managed to install Stable diffusion, but when I get to 8:20 and click "Generate" it keeps telling me "RuntimeError: "LayerNormKernelImpl" not implemented for 'Half'" where the picture shuold be... Any solutions? Thank you very much
please, if someone can help me that would be appreciated
Lo tienes en español?
doesnt work on amd
Hello
How can i get adobe illustrator photoshop and other premium membership for free
I think deforum is much more stable than this workflow and its much simplier to use, I did try this method but deforum works wihtout de-flickering
Hi Jimin
Dropping knowledge! Much appreciated!
🛑The main question remain: how do we make money from it? Can AI animation influence buyer intent?