I was always confused why people uploaded so many pics of themselves to the internet with their name. Anyone who still has photos of themlseves online next to their profile containing every detail of their life has not considered things out. The original advice I received about the internet in 1995 was to remain anonymous, and I have stuck to it. My name still finds no images of me on the internet. AI was always coming, it was only ever that it might be concious that was in question. Doesn't need to be concious to imitate you from your own device in your own voice with full context.
Same. I have a Facebook account just so that I can use it to conveniently log into some other sites, but no pictures of me and not even real name. It's crazy to me that some people would openly broadcast who they are, where they live and what they have just for some compliments from strangers.
Been doing it on purpose for 20 years. I identify as temporarily embarrassed ancestor simulations. I always knew the day the data could be used to create an ancestor simulation. The fools are going to make ancestor simulations of me out of 20 years worth of weaponized autism, in their quest to advertise more efficiently. It's not a bug, it's a feature.
Its very good to see you guys are aware. But his system is build in a way that the few who choose to opt out are of no concern...the next generation is ready to fill that gap!
After lots of troubleshooting and installing, it works. Thank you. Last issue is that I cannot find where I set the number of frames for the video output. I have a 975 frame source video, but the system only renders the first 200 frames. Any ideas why please?
Thanks for the video.I have a question.I installed it but it doesn't work.As it turns out need to install "Detectron2". FFMPEG and Python directly to the C drive. But where should Detectron2 be installed?
i failed installing vid2densepose, with error Failed building wheel for detectron2 and Could not build wheels for detectron2, which is required to install pyproject.toml-based projects , can anyone help, thanks
Everytime i run the gradio, i get some weird huggingface error - huggingface_hub.utils._validators.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name. Went to root folder magic-animate and ran the git commands, what could I be doing wrong?
Thank you! I got error in Comfy UI dont know how to solve Error occurred when executing IPAdapterApply: Error(s) in loading state_dict for Resampler: size mismatch for proj_in.weight: copying a param with shape torch.Size([768, 1280]) from checkpoint, the shape in current model is torch.Size([768, 1024]). and I always got error on installing visual studio on Pinokio
Hello Thanks for the videos ! I'm trying to make Reposer_Plus_BG.json work but got stuck with the error message below. I'm out of ideas :/ Could you help please ? Thanks "Error occurred when executing IPAdapter: 'NoneType' object has no attribute 'patcher'
Sounds like you have a very old version of that custom node installed. I would start by updating everything to the current version, and then following the troubleshooting section.
Set-ExecutionPolicy : Windows PowerShell updated your execution policy successfully, but the setting is overridden by a policy defined at a more specific scope. Due to the override, your shell will retain its current effective execution policy of Bypass.
i also failed. the file "frpc_windows_amd64_v0.2" is recognized as a virus. I do not trust this, so I do not whitelist it. what is it? you should scan it on virustotal. - many scanners are showing the file as virus. @@theamazonclub
Very poor results ( It takes a very long time to generate just for 4 seconds animation. And there can be no question of replacing the tik-tok Dancers. Be realistic!
thanks for guide but maybe next time choose a better reference image poor ai has to handle diffrent proportions and sizes no wonder the output looks weird
It seems I wake up to some awesome new toy to tinker with every day lately, haha! This is certainly going to keep me busy a bit. Thanks much!
You and me both!
Egads Pinokio is a game-changer for my questionably capable self. Wouldn't have learned about it without you @Nerdy Rodent . Very much appreciated!
Easy-peasy, now, isn’t it 😉
any problems? jsut discovered it today
'run this one locally..' you devil, glad ure walking around the earth at the same time as me
Class is in as always, domo arigato Mr.Rodento
I was always confused why people uploaded so many pics of themselves to the internet with their name. Anyone who still has photos of themlseves online next to their profile containing every detail of their life has not considered things out. The original advice I received about the internet in 1995 was to remain anonymous, and I have stuck to it. My name still finds no images of me on the internet. AI was always coming, it was only ever that it might be concious that was in question. Doesn't need to be concious to imitate you from your own device in your own voice with full context.
Same. I have a Facebook account just so that I can use it to conveniently log into some other sites, but no pictures of me and not even real name.
It's crazy to me that some people would openly broadcast who they are, where they live and what they have just for some compliments from strangers.
Bcs we are unvaluable peoples in 3rd world country guys :/
Been doing it on purpose for 20 years. I identify as temporarily embarrassed ancestor simulations.
I always knew the day the data could be used to create an ancestor simulation.
The fools are going to make ancestor simulations of me out of 20 years worth of weaponized autism, in their quest to advertise more efficiently.
It's not a bug, it's a feature.
I choose not to live in fear ❤
Its very good to see you guys are aware. But his system is build in a way that the few who choose to opt out are of no concern...the next generation is ready to fill that gap!
Thanks Nerdy!!!! 🎉🎉🎉
😃
I figured it could be used with comfyui. Very cool. Any difference in the VRam usage in comfyui vs the standard app?
why shows me runtime error in github
??
Is it possible to adjust the densepose thickness of the body to more accurately portray the thing you want to animate? Also, does it do face movement?
No, there isn’t really a face - it’s just a pose
After lots of troubleshooting and installing, it works. Thank you. Last issue is that I cannot find where I set the number of frames for the video output. I have a 975 frame source video, but the system only renders the first 200 frames. Any ideas why please?
The video input is control by the load video node. There you can skip and limit frames.
@@NerdyRodent Got it. Many thanks. You are the rodent king
como es que se ejecuta python3 -m demo.gradio_animate?, en donde tengo que ejcutar eso?, o hacer un archivo ejecutable para usar gradio?
Thank you for another great video. In the comfy UI section I am struggling to understand where you got the 3 ControlNet Model safetensors.
The resources section at the top has links to a load of control nets 😀
towards the end i couldnt tell if i was watching a generated video or a real tik toker
Ikr!
How about for MacOS?
Very good video thank you 😊
Thank you too!
Thanks for the video.I have a question.I installed it but it doesn't work.As it turns out need to install "Detectron2". FFMPEG and Python directly to the C drive. But where should Detectron2 be installed?
You can install in the location of your choice - just like I did 😀
i failed installing vid2densepose, with error Failed building wheel for detectron2 and Could not build wheels for detectron2, which is required to install pyproject.toml-based projects , can anyone help, thanks
Thanks a lot!!! i loved your workflow btw, is it possible to share with us?
can we run it 3080ti 8gb? (ignoring the performance)
8GB may be pushing it a bit, but I think some people have said it works!
I kind of thought Tik tok dancers were already obsolete..
:shocked rodent emoji:
Everytime i run the gradio, i get some weird huggingface error - huggingface_hub.utils._validators.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name. Went to root folder magic-animate and ran the git commands, what could I be doing wrong?
You need to edit your yaml config file to point to your checkpoints
How we can use it in colab plz and who i can use comfyui plllz
What’s the issue with the colab?
@NerdyRodent I don't know how I can use it on colab cus I'm not good in programming so if don't mind we need Tutorial
@@NerdyRodent and comfyui also
Use this video for the comfy one, and the huggingface space if you’ve not actually used colab before 😀
do you have an option to do the same with Colab?
Certainly looks like from the colab in the video!
Thank you!
I got error in Comfy UI dont know how to solve
Error occurred when executing IPAdapterApply:
Error(s) in loading state_dict for Resampler:
size mismatch for proj_in.weight: copying a param with shape torch.Size([768, 1280]) from checkpoint, the shape in current model is torch.Size([768, 1024]).
and I always got error on installing visual studio on Pinokio
Check the ComfyUi troubleshooting tips for more info!
Run in my RTX 3060 12 GB?
我8G跑不起来
Hello
Thanks for the videos ! I'm trying to make Reposer_Plus_BG.json work but got stuck with the error message below. I'm out of ideas :/ Could you help please ? Thanks
"Error occurred when executing IPAdapter: 'NoneType' object has no attribute 'patcher'
Sounds like you have a very old version of that custom node installed. I would start by updating everything to the current version, and then following the troubleshooting section.
@@NerdyRodentthanks, but I already used the "update all buton". Where is the troubleshooting section plz ?
The troubleshooting section is right at the top, just underneath the overview - github.com/nerdyrodent/AVeryComfyNerd#troubleshooting
@@NerdyRodentThanks I read it. I reinstalled IPAdatper plus, still the same issue.
Check the GitHub page for that custom node - maybe someone else has a similar issue open?
Hey nice artwork! Just want to ask that are you interested in our new ai generation product? Maybe I can send you beta version to you~
Dam! thats really bad man to be honest but you did good to help us realize not to try for now
AI is going to destroy youtubers talking about AI destroying tiktokkers.
Set-ExecutionPolicy : Windows PowerShell updated your execution policy successfully, but the setting is overridden by
a policy defined at a more specific scope. Due to the override, your shell will retain its current effective
execution policy of Bypass.
That’s a Microsoft windows error. You’ll need to follow the instructions in the order provided on the GitHub page
I still dont know how to install it on windows, i tried and i failed
Even with pinokio? You’ll need an Nvidia GPU though. There’s always colab and huggingface for easy mode!
@@NerdyRodent Tried separate windows install and through Pinokio - always failed :(
i also failed. the file "frpc_windows_amd64_v0.2" is recognized as a virus. I do not trust this, so I do not whitelist it. what is it? you should scan it on virustotal. - many scanners are showing the file as virus.
@@theamazonclub
Obsolete.
Take the twitter name change seriously when Elon gives the same courtesy to trans people.
Really, as if we need another TicTok dance video aaaaggggghhhh
Very poor results ( It takes a very long time to generate just for 4 seconds animation. And there can be no question of replacing the tik-tok Dancers. Be realistic!
But I want to over react!! 😆
thanks for guide but maybe next time choose a better reference image poor ai has to handle diffrent proportions and sizes no wonder the output looks weird
🫶
💙
Thats bad dislike we want these both in a1111
DensePose Estimation!
Ikr!
is that your voice or is it AI too?
I’m not sure myself at this point!