I might not even use this tool but I gotta say, love the way you explain everything from the installation of conda or any of the pre requisites . Not everyone has the patience or even the slightest of sense to do this, most youtubers diving into these AI tools are programmers themselves and end up assuming the entire world is as well. So I absolutely appreciate how you've dedicated your time in explaining each and every detail.
@@lucifer9814 ha. I got a whole comment section deleted by a youtube with rodent in his name. I've been on youtube from the beginning, never ever had an entire comment section nuked. I said the same thing and left a comment, rodent didnt agree and said he explains for all... Then the comments started flying in most agreeing with me. I think sum ppl dont even know they are talking super advanced, maybe it comes so natural they assume EVERYONE is on same level.
@@lucifer9814 good on ya. It winds me. Example... "Ok let's create a repository in the cache section, open up cmd prompt make sure you got all the components, python 9.1.0.7 , kittie ver 11.47, (not ver 11.47 old version) alligator ver 10.67543 LM, set all variables to bat file and make sure you tick the binary only option in the windows key prompt pop up. Now set to c:users/programfiles/cmdcenter/oneclick/errormessage/igiveup
nice buddy. you have a very clear and concise way of presenting and explaining tasks without sounding too techy. that is a gift. the python and conda installation instructions were absolutely on the money. so many of these pretenders could take a lesson or 2 from you. subbed.
we are slowlu but surely getting closer and closer to fully generating entire comic books with AI, seeing how we get much more control over our generations is insane!
Full movies, really. First we need an AI that's good enough at fiction The current ones available are just not there yet. I'm sure if they train something specifically for it, we could get there now, I feel like it's probably going to be a few more years till we're getting actual good stories out of AI.
@@jaredf6205after that if you don't have the brain to know how to create fiction that's your problem, what's interesting is the tools for those who already have projects in order to bring them to life
This is super useful. Even though it was possible to do this with other AI's or just goof around with the liquefy tool in Photoshop, this one is most convenient, despite having garbage controls. I went and put a few thumbnails on my channel through it, taking regular people's faces and turning them goofy. I can tell that my click rate increase 10% as a result, since "clickbait" surprised expressions tend to get more views. Highly recommend it.
I get a black square block over the entire face and then nothing happens until I reload another image, which just leads to same thing. Anywhere I click and drag the mouse, it just shows a black block over the face.
Hey there! First of all, thank you for your video. It is very helpful for people like me who are not very versed in informatics. I just wanted to know if there is a way to manually select the installation folder for all the dependencies, because right now I'm struggling with space on my C: drive. Thanks in advance and keep up the good work.
It is kinda limiting, if you had mouth fenome shapes you could do short animations of a character speaking. Not being able to alter eye direction seems like thats the biggest Achilles heel of the program. With everything else it can do you would think you could at least make a character look left right up or down.
I make a sort of tv show for me and my friends, but since there is no budget I'm not going to buy a greenscreen to shoot the scenes irl and I instead use 2d cutouts of me, some friends and some stock photo guys. The big downside of this was that for most character I only ever had one expression, maybe a couple at most. A tool like this will greatly improve the storytelling in season two
the dependencies and requirements that you have installed in this video are stored on our pc or on virtual environment? do we need to delete those to clear up the space?
They both have things each lack. But it looks there is still a problem if your picture has long hair. The hair doesnt change with the head turning. I am using it right now on anime characters.
Thanks a lot! Usually this stuff works with no issues but this is the first time I've encountered an error: The packages start installing normally and then I get a red wall of text: ERROR: Ignored the following yanked versions: 2.8.0, 3.0.7, 3.1.2, 3.13.1, 4.7.0 ERROR: Could not find a version that satisfies the requirement gradio==5.0.0b4 (from versions............ ......... ......... (some version numbers in between ......... ERROR: No matching distribution found for gradio==5.0.0b4 Would you know how to fix this?
I installed Facepoke with Pinokio, which took about 45 minutes and 30gb of disk space. However, after loading up a picture and moving any facial area with the mouse, it shows a black block over the entire area. The main window also doesn't allow the entire image to be moved around (image is locked in place with no scroll bars).
Just tried it. multiple images. the face landmarks would show, but they would never respond or make any changes. The image wouldn't move. I couldn't click and drag and make anything react.
Dear friend, could you teach how to make a configuration to be able to animate several faces with different reference videos each, this can be done with LivePortrait plus other resources, but those tutorials that I have seen, I have not seen a detailed explanation of how to do it, for that's it, I would like you to make that video, thank you
Cannot find "edit the system environment variables" in my pc edit: Yuo can get into the setting via Win + R and type sysdm.cpl > card Advandced > Enviromanet Variables
Not sure why the toggle show highlight landmark on the face thing doesn't work for me despite I habe checked the option. ill try to troubleshoot. Any tips from anyone appreciated.
Games like Dust Tales of the wired west, old point and click adventure games from the 90s with a set of expressions by a real actor with voice acting as you're talking to people in the game. This would allow you to do that. Walk up to a 3d figure, enter a conversation and be greeted by a photo that talks in response to your prompts with voiced lines. You can even manipulate the voice of a person using AI, so you could voice act the entire thing yourself assuming you have trained models of different voices. If you have knowledge of painting you might even be able to use it on semi realistic drawings. This has great potential in some genres of video games, outside of the video games. Very limited usage.
@@martian8987What those old games had were a set of expressions for (eyes, mouth, eyebrows) and a base head tilt/roll/yaw, each of which created a new set of expressions. So you'd probably use this for the base, so that the hair neck and non facial features got the proper position. Then you'd deepfake your own face with the character's face, and use that for facial expressions. Extract frames from the video, photoshop them onto the different expressions. Probably no more than 40 frames per character for a range of emotions. You create an expression class consisting of a list of images that are iterated through after the player picks a dialogue choice. In Dust there were about 15 characters. So about 600 images required, most of those being moving mouths. Definitely doable, it's however a lot of work for a niche feature. Especially the voice acting. Because you would need to create 15 voice models, one of those voices takes 7 hour to train. And you couldn't just take a random nicely sounding voice because you don't have the right to it. So you'd have to merge voice models together for a voice that doesn't exist. Doing that for 15 voice models, at least 100 hours. Then the dialogue itself, an interesting setting and an engaging plot. From start to finish about a year. positive thing is you don't have to animate a lot of 3d models or build advanced behavior. Puzzle games relying on conversations are incredibly easy to code.
@@DanIel-fl1vc I think for making Mods, this would be interesting (probably too much code bloat, but hey, its free) - so we can create Might and Morrowind, with Dagoth Ur and Johnny Depp (could even take actors from movies - would be pretty funny). Hypothetically (which means never).
For the love of God! Why don't these huggingnerds just put a freaking download & install freaking button!!? If it's loaded with options and versions I can answer questions like location I want it in or whatever. I swear! I remember a day and age you click a few times and you're done. All that nerdy command window and secret hacking codes and you STILL can't make them stick out their tongue! Fu'all'dat'sht BTW, these are the best install instructions I've seen yet for dealing with cLoNiNg RePoSiToRiEs
So, I am stuck with these error and I can't find any seclusion on google: ERROR: could not find a version that satisfies the requirement gradio==5.0.0b4
Doesn't work for me. I'm using Windows 11 with a Ryzen 5 5600h 3.3Ghz and 16gb of RAM, so obviously the setup was not configurated properly. Mine was installed by Pinokio, which took 45 minutes and well over 30gb of disk space, almost as much as a good modern video game! The yellow outlines over facial areas appear like eyebrows, eyes, etc.. but clicking and dragging the mouse just makes a black block appear over the entire face. Then it does nothing. Regardless, I can't imagine all of this space being needed to simply move a head or face around. As a matter of fact, I know it doesn't need that much space to achieve something almost trivial. That in itself is pretty ridiculous.
I tried three time download the list and i am getting an error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> [23 lines of output] Traceback (most recent call last): File "", line 9, in ModuleNotFoundError: No module named 'pip' During handling of the above exception, another exception occurred: File "", line 11, in ModuleNotFoundError: No module named 'pip' [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> See above for output. note: This error originates from a subprocess, and is likely not a problem with pip.
🔵Intelligence is compression of information.☑This is one of the most useful videos I believe I have ever watched on TH-cam.☑ "I absolutely love this channel! The content is always top-notch, informative, and crafted with such care. Every video offers valuable insights and presents topics in a clear and engaging way. You can really see the effort and dedication put into every detail, consistently delivering useful and interesting information. Thank you for creating such an amazing space to learn and be inspired. Keep up the fantastic work!☑
I ran in to a problem with this C:\Users\xxxxx\OneDrive\Desktop>conda activate facepoke CondaError: Run 'conda init' before 'conda activate' Is there anything i can do to fix this and continue? Thank you.
Thanks to our sponsor Abacus AI. Try their new ChatLLM platform here: chatllm.abacus.ai/?token=aisearch
I might not even use this tool but I gotta say, love the way you explain everything from the installation of conda or any of the pre requisites . Not everyone has the patience or even the slightest of sense to do this, most youtubers diving into these AI tools are programmers themselves and end up assuming the entire world is as well. So I absolutely appreciate how you've dedicated your time in explaining each and every detail.
@@lucifer9814 ha. I got a whole comment section deleted by a youtube with rodent in his name.
I've been on youtube from the beginning, never ever had an entire comment section nuked.
I said the same thing and left a comment, rodent didnt agree and said he explains for all... Then the comments started flying in most agreeing with me.
I think sum ppl dont even know they are talking super advanced, maybe it comes so natural they assume EVERYONE is on same level.
@@armondtanz lol I've spammed quite a few such tubers on their videos when they give a vague explanation about their stuff
@@lucifer9814 good on ya. It winds me.
Example...
"Ok let's create a repository in the cache section, open up cmd prompt make sure you got all the components, python 9.1.0.7 , kittie ver 11.47, (not ver 11.47 old version) alligator ver 10.67543 LM, set all variables to bat file and make sure you tick the binary only option in the windows key prompt pop up.
Now set to c:users/programfiles/cmdcenter/oneclick/errormessage/igiveup
How do I do this on mac bro its so frustrating and time consuming
"You can't make her cross-eyed, you can't stick her tongue out, you can't do any of those things. There's no tongue involved here."
_clicks off video_
did you mean "jack"
so you cant do the most fun things meh
yall are down nasty :/ why can't we just enjoy ai tools for the sake of them being interesting?
@@276-Por que está tão sério?
🥵
Good enough. Still WOULD
' cross - eyed ', ' tongue out ', Hmmmm.... I wonder why you specifically choose those words 🤔
🤔🤔🤔
@@theAIsearchcoomer
😂
@@theAIsearchbased
🤔🤔🤔🤔🤔
This was an eye-opener, thanks
you're welcome, and nice pun!
That kind of pun
is frowned upon.
but not an eye-crosser,sadly
Sadly I can't see this joke for the second time
nice buddy. you have a very clear and concise way of presenting and explaining tasks without sounding too techy. that is a gift. the python and conda installation instructions were absolutely on the money. so many of these pretenders could take a lesson or 2 from you. subbed.
thanks!
we are slowlu but surely getting closer and closer to fully generating entire comic books with AI, seeing how we get much more control over our generations is insane!
U right
What issues prevent someone creating a comic right now? I thought AI generated images were already On par with the best artists?
@@Dataism AI image generators still don't give users character consistency option, so the character looks different in each generation so not usable
Full movies, really. First we need an AI that's good enough at fiction The current ones available are just not there yet. I'm sure if they train something specifically for it, we could get there now, I feel like it's probably going to be a few more years till we're getting actual good stories out of AI.
@@jaredf6205after that if you don't have the brain to know how to create fiction that's your problem, what's interesting is the tools for those who already have projects in order to bring them to life
3:31 The only necessary feature is missing
1:30 You are the first youtuber in a long time who pronounces "Et Cetera" correctly, instead of saying "Ek Cetera"
making ahegao faces for fun
😏😏😏
Kkkkkkkkkkkkk melhor ideia 💡
based
Yes
AI PH is on the way
This is super useful. Even though it was possible to do this with other AI's or just goof around with the liquefy tool in Photoshop, this one is most convenient, despite having garbage controls. I went and put a few thumbnails on my channel through it, taking regular people's faces and turning them goofy. I can tell that my click rate increase 10% as a result, since "clickbait" surprised expressions tend to get more views. Highly recommend it.
Finally, I can make anyone into a Super Mario 64 intro screen.
You always had the power, you just had to believe!
Face poke is not working for me. I’m dragging my mouse around but nothing happens and I’m making sure parts of the face are highlighted
I tried with Safari and it didnt work, but it worked when I switched to Firefox.
@@densakai I tried with chrome and brave and didn’t work but thanks for letting me know
do not works for me too, 4 browsers...none
I get a black square block over the entire face and then nothing happens until I reload another image, which just leads to same thing. Anywhere I click and drag the mouse, it just shows a black block over the face.
It gets scary! Thanks for the videos sir!🔥
you're welcome!
Only scary if you let it be scary. Have faith… what is the worst that could happen? 😂
Hello is there a way to local install Expression Editor??
atleast I am not required to learn manga editing and animation, saved 40 hours of my time thx
Bro, you can 1-click install this using pinokio.
thanks for sharing!
What is it
@@lucagenovese7207 pinokio is a one-stop app that can install a lot of AI stuffs
Make the tutorial bro
bro thank you
Thankyou, please know that you are helping us, you deserve all the recognition you get
Hey there! First of all, thank you for your video. It is very helpful for people like me who are not very versed in informatics. I just wanted to know if there is a way to manually select the installation folder for all the dependencies, because right now I'm struggling with space on my C: drive. Thanks in advance and keep up the good work.
It is kinda limiting, if you had mouth fenome shapes you could do short animations of a character speaking. Not being able to alter eye direction seems like thats the biggest Achilles heel of the program. With everything else it can do you would think you could at least make a character look left right up or down.
I was pjayong with these a few hours ago, very useful for what im working on
Great to hear!
Lol what is bro working on
Você é um homem determinado, por acaso estava trabalhando usando apenas uma mão?
@@Joao-1114 🤣LOL i'm dying out of laughter
I make a sort of tv show for me and my friends, but since there is no budget I'm not going to buy a greenscreen to shoot the scenes irl and I instead use 2d cutouts of me, some friends and some stock photo guys. The big downside of this was that for most character I only ever had one expression, maybe a couple at most.
A tool like this will greatly improve the storytelling in season two
I love that you keep it super beginner friendly 😊
Journalists best friend for photos of people they don't like
lol
Can it do Ahegao? It's for research.
the dependencies and requirements that you have installed in this video are stored on our pc or on virtual environment? do we need to delete those to clear up the space?
This is pretty cool, I think it did a great job.
yes!
Would you be able to share the specifications of the computer that was used?
They both have things each lack. But it looks there is still a problem if your picture has long hair. The hair doesnt change with the head turning. I am using it right now on anime characters.
Thanks a lot! Usually this stuff works with no issues but this is the first time I've encountered an error: The packages start installing normally and then I get a red wall of text:
ERROR: Ignored the following yanked versions: 2.8.0, 3.0.7, 3.1.2, 3.13.1, 4.7.0
ERROR: Could not find a version that satisfies the requirement gradio==5.0.0b4 (from versions............
.........
......... (some version numbers in between
.........
ERROR: No matching distribution found for gradio==5.0.0b4
Would you know how to fix this?
your video is very helpful, thank you very much❤❤❤
I installed Facepoke with Pinokio, which took about 45 minutes and 30gb of disk space. However, after loading up a picture and moving any facial area with the mouse, it shows a black block over the entire area. The main window also doesn't allow the entire image to be moved around (image is locked in place with no scroll bars).
Thanks for a great video. Could you specify if downloads are Windows/Mac/Linux only - just so us Mac user know when we can step in or out? Thanks
from their github, it seems this only works with a nvidia cuda gpu. hope that helps
How does it work for a phone? Is there a downloadable version for a phone?
Did u do a videos on face swap?
is there something similar but for changing poses?
The expression editor focuses on the face but don't integrate the result back to the full image?
This is all fine but can you make a video of these movements or what good is this for?
see my liveportrait tutorial to make videos of this th-cam.com/video/uyjSTAOY7yI/w-d-xo.html
Does it have to be installed or can it be used directly from the computer?
Just tried it. multiple images. the face landmarks would show, but they would never respond or make any changes. The image wouldn't move. I couldn't click and drag and make anything react.
You made Blue Steel Musk. 💙
Thanx for the sharing. This is AMAZING technology. Peace & love
What's the quality like of the modified image? Because when I use LivePortrait the output image resolution is so low after doing any modifications.
This is how future games will work..
interesting thought
OMG thanks!
Is there a way to locally install Expression Editor?
It only works with linux ?? i tried it on windows , with opera and chrome and it doesnt work
Same here.
i'm using windows and it works. try installing here: pinokio.computer/item?uri=github.com/pinokiofactory/facepoke
@@theAIsearch thanks but I meant does not work online on the browser. I'll try to install it locally
is there a body shaping software.
Dear friend, could you teach how to make a configuration to be able to animate several faces with different reference videos each, this can be done with LivePortrait plus other resources, but those tutorials that I have seen, I have not seen a detailed explanation of how to do it, for that's it, I would like you to make that video, thank you
What's the GPU memory requirement for running it locally?
That is a very nice thumbnail I must say.
hello miniconde requires video board?
friends, maybe someone knows something similar for turning a character? I have tried “runway” - many artifacts
This is totally cool😂😅
I like expression editor much more but how do I run that one locally?
just install liveportrait. it has those sliders: github.com/KwaiVGI/LivePortrait/blob/main/assets/docs/changelog/2024-08-06.md
is nightshade a liable way to protect the photos with my face for the cases of this being used horribly against me?
It would be amazing if we can import the rigging for Moho or sim softwares
Can we install expression editor locally?
This is like the GMod face poser tool.
the expression editor seem superior, sadly tho im not finding a way to run it locally so it automatically loses the race for me
Great tutorial THX!!
No problem!
nothing moves when i move my mouse, why is that?
did you click & drag?
@@theAIsearch of course
@@HE572-hp3ur do you see any errors
@@theAIsearch no i see the yellow line but when i click and move my mouse nothing moves
@@theAIsearch hello today just worked suddenly thanks
Very helpful , I finally managed to install it thanks to your comprehensive tutorial !
For some reason when I add path it still gives me: conda is not recognized or external command.
did you start a new cmd first
@@theAIsearch It helped, thanks
This channel is so good for updates
Is expression editor work on phone or no ?
Cannot find "edit the system environment variables" in my pc
edit: Yuo can get into the setting via Win + R and type sysdm.cpl > card Advandced > Enviromanet Variables
fascinating and scary at the same time
Not sure why the toggle show highlight landmark on the face thing doesn't work for me despite I habe checked the option. ill try to troubleshoot.
Any tips from anyone appreciated.
I wrote about the browser version. Will check tomorrow in various browsers and extensions
Finally I can make myself look normal
Can it make same with the video i have?
I am testing this PokeFace in a new Mac and it doesn´t do anything :(
i think you need a cuda gpu for this
very, very cool, another subscriber
is there a whole body version?
Games like Dust Tales of the wired west, old point and click adventure games from the 90s with a set of expressions by a real actor with voice acting as you're talking to people in the game. This would allow you to do that. Walk up to a 3d figure, enter a conversation and be greeted by a photo that talks in response to your prompts with voiced lines. You can even manipulate the voice of a person using AI, so you could voice act the entire thing yourself assuming you have trained models of different voices. If you have knowledge of painting you might even be able to use it on semi realistic drawings. This has great potential in some genres of video games, outside of the video games. Very limited usage.
yes, finally I can remake Doom and Might and magic 6
@@martian8987What those old games had were a set of expressions for (eyes, mouth, eyebrows) and a base head tilt/roll/yaw, each of which created a new set of expressions.
So you'd probably use this for the base, so that the hair neck and non facial features got the proper position.
Then you'd deepfake your own face with the character's face, and use that for facial expressions. Extract frames from the video, photoshop them onto the different expressions.
Probably no more than 40 frames per character for a range of emotions. You create an expression class consisting of a list of images that are iterated through after the player picks a dialogue choice.
In Dust there were about 15 characters. So about 600 images required, most of those being moving mouths. Definitely doable, it's however a lot of work for a niche feature. Especially the voice acting. Because you would need to create 15 voice models, one of those voices takes 7 hour to train.
And you couldn't just take a random nicely sounding voice because you don't have the right to it. So you'd have to merge voice models together for a voice that doesn't exist. Doing that for 15 voice models, at least 100 hours. Then the dialogue itself, an interesting setting and an engaging plot.
From start to finish about a year. positive thing is you don't have to animate a lot of 3d models or build advanced behavior. Puzzle games relying on conversations are incredibly easy to code.
@@DanIel-fl1vc I think for making Mods, this would be interesting (probably too much code bloat, but hey, its free) - so we can create Might and Morrowind, with Dagoth Ur and Johnny Depp (could even take actors from movies - would be pretty funny). Hypothetically (which means never).
For the love of God! Why don't these huggingnerds just put a freaking download & install freaking button!!? If it's loaded with options and versions I can answer questions like location I want it in or whatever. I swear! I remember a day and age you click a few times and you're done. All that nerdy command window and secret hacking codes and you STILL can't make them stick out their tongue! Fu'all'dat'sht
BTW, these are the best install instructions I've seen yet for dealing with cLoNiNg RePoSiToRiEs
we all know where this is going
This is amazing
So, I am stuck with these error and I can't find any seclusion on google: ERROR: could not find a version that satisfies the requirement gradio==5.0.0b4
Doesn't work for me. I'm using Windows 11 with a Ryzen 5 5600h 3.3Ghz and 16gb of RAM, so obviously the setup was not configurated properly. Mine was installed by Pinokio, which took 45 minutes and well over 30gb of disk space, almost as much as a good modern video game! The yellow outlines over facial areas appear like eyebrows, eyes, etc.. but clicking and dragging the mouse just makes a black block appear over the entire face. Then it does nothing. Regardless, I can't imagine all of this space being needed to simply move a head or face around. As a matter of fact, I know it doesn't need that much space to achieve something almost trivial. That in itself is pretty ridiculous.
Much fun🎉
I tried three time download the list and i am getting an error
× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> [23 lines of output]
Traceback (most recent call last):
File "", line 9, in
ModuleNotFoundError: No module named 'pip'
During handling of the above exception, another exception occurred:
File "", line 11, in
ModuleNotFoundError: No module named 'pip'
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error
× Getting requirements to build wheel did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
Please upload the video "Microsoft Copilot Voice Guide
thanx, liked it a lot 🤙
Thanks 😊
🔵Intelligence is compression of information.☑This is one of the most useful videos I believe I have ever watched on TH-cam.☑
"I absolutely love this channel! The content is always top-notch, informative, and crafted with such care. Every video offers valuable insights and presents topics in a clear and engaging way. You can really see the effort and dedication put into every detail, consistently delivering useful and interesting information. Thank you for creating such an amazing space to learn and be inspired. Keep up the fantastic work!☑
Thanks🎉
you're welcome!
I ran in to a problem with this
C:\Users\xxxxx\OneDrive\Desktop>conda activate facepoke
CondaError: Run 'conda init' before 'conda activate'
Is there anything i can do to fix this and continue? Thank you.
enter conda init first
Can i install these on android
wow, cool tool,thanks for sharing
think I’m going to try animating an AI image using it.
why Facepoke not working for me ? i upload the photo and marker appears but face just not change
i think it only works on a linux environnement
@@fazeone1212 how? its online i think its not related to specific operation system
do you see any error
@@theAIsearch No , no errors 🤔
@@mostafamito2503 same , i can load the picture but i can't drag to manipulate the image , its not working
FacePoke not working now, any idea why?
Hiw can i download facepoke in android 😢😢please
Can I use for anime chahters?
How we use this tool for flim making?
see my video on liveportrait. you (or anyone else) can act out the expressions, and it'll map it onto a character in the video
Says no NVIDIA driver on your system
thats amazing, i expect the same control for posing
Does it work with animals?
You are smart and unique