I got a question. After making 1 , I tried to make another using a different photo, but when I get to the last part (animate) the generated.mp4 file is still the one I did before. Even if i delete it nothing comes up
I did everything yet when the video is generated, it's not on the left for me to download? There's the three folders but there's no "generated.mp4", even after refreshing
for those who can't find generated.mp4 do this 2. Then you replace this line in the Perform image animation part: replace "imageio.mimsave('../generated.mp4', [img_as_ubyte(frame) for frame in predictions])" for "imageio.mimsave('/content/gdrive/My Drive/first-order-motion-model/generated.mp4', [img_as_ubyte(frame) for frame in predictions])" 3. Now you can find the video in your drive, in the folder "first-order-motion-model" that you created before, with the name "generated.mp4". it was @ Thành Vinh Entertainment who found go tank him
What worked for me is removing the image you want to make into the meme and place it back in the folder again. After that, reload the script page and redo everything, then the generated.mp4 should work.
I entered in my authorization code by pressing enter and all that, but is keeps saying --------------------------------------------------------------------------- TIMEOUT Traceback (most recent call last) in () 1 from google.colab import drive ----> 2 drive.mount('/content/gdrive') 4 frames /usr/local/lib/python3.6/dist-packages/pexpect/expect.py in timeout(self, err) 142 exc = TIMEOUT(msg) 143 exc.__cause__ = None # in Python 3.x we can use "raise exc from None" --> 144 raise exc 145 146 def errored(self): TIMEOUT: searcher: searcher_re: 0: re.compile('google.colab.drive MOUNTED') 1: re.compile('root@762d47c518a0-43a474a6a8474fff88d80de63beaac54: ') 2: re.compile('(Go to this URL in a browser: .*)$') 3: re.compile('Drive File Stream encountered a problem and has stopped') 4: re.compile('drive EXITED') searcher: searcher_re: 0: re.compile('google.colab.drive MOUNTED') 1: re.compile('root@762d47c518a0-43a474a6a8474fff88d80de63beaac54: ') 2: re.compile('(Go to this URL in a browser: .*)$') 3: re.compile('Drive File Stream encountered a problem and has stopped') 4: re.compile('drive EXITED') How do I fix this?
go to top right where it shows (ram/disk), click the drop down menu, manages sessions, and terminate the session, if it still persists repeat but this time close the tab after you terminate and the browser and relog
I've done this once but when trying to do a second one, it says "Cannot connect to GPU backend" but when I click "connect without GPU" it infinitely loads. any way to fix this so I can use my GPU?
Bro, I just find the soluttion. The limit is on your google account that you use to iniciate the code. If u change that account to another one the limit will vanished.
I always got stuck at 2:06 with "all the input arrays must have same number of dimensions, but the array at index 0 has 2 dimension(s) and the array at index 1 has 3 dimension(s)"
Getting the lip syncing right is difficult. Looks much better on this video than on what I made even though I did it exactly according to instructions.
go to top right where it shows (ram/disk), click the drop down menu, manages sessions, and terminate the session, if it still persists repeat but this time close the tab after you terminate and the browser and relog
Is it possible to create a full length Baka Mitai video deepfake. using the original lip sync video and a picture. I tried that, it didn't work because it was a large byte size; so I split them into 9 parts, 30 seconds long, still didn't work. Is there a solution to this?
Okay it all seems cool and I'm grateful for this but the part of using specifically vegas pro and 256 res isn't a bit unnecessary? Like you could use any editing software and the configuration you prefer
tried to make one. somehow the colab page looks different and it always has the same error def thumbnail (file): return imageio.get_reader(file, mode=´I´, format=´FFMEPG´).get_next_data()
please make a version of the file on google drive. that way you dont actually have to download just import my internet is like 10kbps i cant download and then upload it somewhere else
Where i found the dame dane file?? i dont found it in the description bro another thing, the links in the description are buged or something because i can't download it cause there are a lot of annoying spam
@@kirigayakazuto843 Bro demo module is basically not found in that repository, why the fk should someone change picture file to png? when its not png they couldn't pass on the checkpoint before
When I mount the google drive which is linking my google drive account it did not show the part where i need to go to the URL browser & copy the authentication code instead it shows the "Mounted at /content/drive, how can I fix this issue?
I managed to make one, but as soon as I want to make another one, I download it but it appears to be the same image I've used for the first one rather than the second image that I have put in even though I did all the steps again just to change it. But it's still persistent on using the previous image. If that makes sense? help anyone?
Also if it don't work when you're loading the video to deepfake it, Replace : driving_video = imageio.mimread('/content/gdrive/My Drive/¤¤¤¤¤¤.mp4') by driving_video = [] reader = imageio.get_reader('/content/gdrive/My Drive/¤¤¤¤¤¤.mp4') try: for im in reader: driving_video.append(im) except RuntimeError: pass
@@happypiak1391 already downloaded and uploded on my drive. The problem is. In the first 3play button, it ks working but when generating the final output, it always says modulenotfounderror
it still not working at the from demo import load_checkpoints generator, kp_detector = load_checkpoints(config_path='config/vox-256.yaml', checkpoint_path='/content/gdrive/My Drive/first-order-motion-model/vox-cpk.pth.tar')
So I was able to make a couple with the same template then when I changed the video, it gave me an error at the checkpoint step, tried it every which way and eventually used the same template and image I already used (and previously worked) and now it’s not working for me at all, anyone know what to do??
I have a problem. The first time it worked but when i want to try it with an other Image the third "codeblock" (idk) says: " Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount("/content/gdrive", force_remount=True). " how do i fix that?
Bruv, can't download the video . "generated.mp4" is not created im done with everything. UPDATE:- The site might glitch it took me a couple of redos but i was able to get the "generated.mp4" in the end .
for people who are having the FileNotFound problem and youre POSITIVE all the labels are correct, make sure the google drive account is the same account you are logged into at the time if you have multiple accounts. took me days to realize i was logged into the wrong account lol, then it worked.
Me: clicks the download link Viruses: Allow us to introduce ourselves Can someone give the normal link to the downloads? I have 2 viruses on my phone because od him
can you make a google drive mirror? it's easier to just import it to google drive than downloading it and uploading it, also i have a shitty pc with no storage to spare
Okay, after trying a few things I found a way, here's the fix for generated.mp4: See the code that says imageio.mimsave(..................)? substitute it for: imageio.mimsave('/content/gdrive/My Drive/first-order-motion-model/generated.mp4', [img_as_ubyte(frame) for frame in predictions]) Now, you just go to the folder in which you put the base image and video, and you will find the video there. it's a bit slow, for some reason, but you can fix it in vegas at least You can subscribe if you want to thank me for some reason
@@fraddyfuzzybear2756 That happened to me too before, what I did was just redo all the steps, I don't think it's a fault of python, it just doesn't find the required stuff OR the cd changes. Do cd again (the second code segment)
I did it, but after making a couple videos, I got an error message when I tried to connect again saying (Cannot connect to GPU backend). Anyone know what this means and how to fix it?
I try to click the "Load driving video and source image" block and this is the error I get: RuntimeError: imageio.mimread() has read over 256 MiB of image data. Stopped to avoid memory problems. Use imageio.get_reader() or memtest=False. I have no knowledge of coding, how do I fix this?
under the play button for that block you'll see something that says 'imageio.mimread(some text in red)', replace the 'imageio.mimread' with 'imageio.get_reader' and run it
*DAME* *DA* *NE* *MEME* *ANDROID* *TUTORIAL:* ruclips.net/video/8PQbehX8NMU/видео.html
*COMMON* *ERRORS* *IN* *MAKING* *DAME* *DA* *NE* *MEME:* ruclips.net/video/7CK8wqsvOwo/видео.html
I got a question. After making 1 , I tried to make another using a different photo, but when I get to the last part (animate) the generated.mp4 file is still the one I did before. Even if i delete it nothing comes up
How do I get it to not go over in memory?
bro help pls
How to fix no module named demo pls pls
Yo the one check points oneis taking a long time it was red before so I tapped it again and it's loading for a long time is this supposed to happen?
*Careful, he's a hero*
no dont beat my like pleasee lol
?
This is a whilesome 100 reddit coin moment whileisme momentreddot gold
The
Link for Midafire is an Ad page?
Gumball
Every human alive need this tips to survive the jungle of quarantine
So true, and to piss of some friends.
You got 473 for now.
700th like
Thank you for the tutorial. I had a lot of fun making one for my friends and I. You are a wonderful person.
damn it's so formal lmao
i think you forgot to put the link to the original source aka me!
bro your the guy lol
thats epic
epiccc
damn thats cool as hell
Lol you should update the description to put him there.. Remember, without him, none of these would be exist. 😏
I did everything yet when the video is generated, it's not on the left for me to download? There's the three folders but there's no "generated.mp4", even after refreshing
try using incognito
same too
@@MirForge incoginto mode is work, thank you
@@flynnriderz-zf k
@@MirForge incognito still don't work help
Steps are pretty straightforward! The folder name and file type and file names matter the most! they have to be exactly the same :)
"first-order-motion-model" I got you homie
for those who can't find generated.mp4 do this
2. Then you replace this line in the Perform image animation part:
replace "imageio.mimsave('../generated.mp4', [img_as_ubyte(frame) for frame in predictions])"
for "imageio.mimsave('/content/gdrive/My Drive/first-order-motion-model/generated.mp4', [img_as_ubyte(frame) for frame in predictions])"
3. Now you can find the video in your drive, in the folder "first-order-motion-model" that you created before, with the name "generated.mp4".
it was @ Thành Vinh Entertainment who found go tank him
works for me thx
directions too hard, whole computer is corrupted
Can you do a meme for me?
What worked for me is removing the image you want to make into the meme and place it back in the folder again. After that, reload the script page and redo everything, then the generated.mp4 should work.
You don't need Vegas Pro
If you don't want to buy a very expensive piece of software, don't. You can use pretty much any video editor
u can download online for free just type vegas pro download free on chrome and you'll find it
@@ianw.8132 That is illegal
Pirating Software, Is illegal
@@Catmangamer you're right
me: clicks links in description
website: YOU WON'T LAST 5 SECONDS PLAYING THIS GAME
Relatable, I know every hot girl lives near to me now thanks for 100 pop up ads
@@mertkaya8097 and won 1 million dollars in just 3 min and 69 seconds
bruh make these tutorials for every famous video meme pls
Uploaded on my birthday, nice.
happy birthday mate
Happy birthday bruh
happy cake day bro
Gotrush thanks bro 👍
Happy birthday
fatal: destination path 'first-order-model' already exists and is not an empty directory. ?
wtf ?
rename the folder to another name, then try clicking play again. good luck ^^
i reloaded and it seem to fix it
it's called first-order-motion-model
I entered in my authorization code by pressing enter and all that, but is keeps saying
---------------------------------------------------------------------------
TIMEOUT Traceback (most recent call last)
in ()
1 from google.colab import drive
----> 2 drive.mount('/content/gdrive')
4 frames
/usr/local/lib/python3.6/dist-packages/pexpect/expect.py in timeout(self, err)
142 exc = TIMEOUT(msg)
143 exc.__cause__ = None # in Python 3.x we can use "raise exc from None"
--> 144 raise exc
145
146 def errored(self):
TIMEOUT:
searcher: searcher_re:
0: re.compile('google.colab.drive MOUNTED')
1: re.compile('root@762d47c518a0-43a474a6a8474fff88d80de63beaac54: ')
2: re.compile('(Go to this URL in a browser: .*)$')
3: re.compile('Drive File Stream encountered a problem and has stopped')
4: re.compile('drive EXITED')
searcher: searcher_re:
0: re.compile('google.colab.drive MOUNTED')
1: re.compile('root@762d47c518a0-43a474a6a8474fff88d80de63beaac54: ')
2: re.compile('(Go to this URL in a browser: .*)$')
3: re.compile('Drive File Stream encountered a problem and has stopped')
4: re.compile('drive EXITED')
How do I fix this?
go to top right where it shows (ram/disk), click the drop down menu, manages sessions, and terminate the session, if it still persists repeat but this time close the tab after you terminate and the browser and relog
It says It takes 30 hours for the files to go onto google drive 😭
damn internet
mine says 80 hours 😭
damn what y'alls internet service providers
Cry globe
it takes me 10 minutes LOL
i changed my pic file to png and my video to mp4, nut the play button is still red and there is zig zag line under my video code
I’m stuck at the part where you load in the image and video and it keeps creating a red play button. I named the files properly and everything
did u download the 1G
SmileeS vR Idk
@@flee2573 what is 1G?
I figured it out use, search the easiest deepfake tutorial with a single image and its the first one, it works perfectly.
FileNotFoundError
edit: nvm fixed it, you should note that the picture file has to be png
Underrated af
ty man
or just change it to .jpg
i changed it to 02.png but it doesnt work
@@sanemi3103 of course it doesn't
Renamed it to png didnt work
I've done this once but when trying to do a second one, it says "Cannot connect to GPU backend" but when I click "connect without GPU" it infinitely loads. any way to fix this so I can use my GPU?
x2 bro, lets keep searching
Bro, I just find the soluttion. The limit is on your google account that you use to iniciate the code. If u change that account to another one the limit will vanished.
Voltrion Surely there’s another way to do it
I always got stuck at 2:06 with "all the input arrays must have same number of dimensions, but the array at index 0 has 2 dimension(s) and the array at index 1 has 3 dimension(s)"
yeah same with me bro
ruclips.net/video/7CK8wqsvOwo/видео.html
Getting the lip syncing right is difficult. Looks much better on this video than on what I made even though I did it exactly according to instructions.
fatal: destination path 'first-order-model' already exists and is not an empty directory. how do i fix this?
rename the folder to another name, then try clicking play again. good luck ^^
it’s first-order-motion-model
go to top right where it shows (ram/disk), click the drop down menu, manages sessions, and terminate the session, if it still persists repeat but this time close the tab after you terminate and the browser and relog
i cant do the part at 2:06 my file name is right everything looks good it gives me a red play button and the red squiggle under my photo
same
you have to put your files into a folder named "first-order-motion-model"
resize the pic to 256x256 and make sure it's .png
I figured it out use, search the easiest deepfake tutorial with a single image and its the first one, it works perfectly.
Is it possible to create a full length Baka Mitai video deepfake. using the original lip sync video and a picture. I tried that, it didn't work because it was a large byte size; so I split them into 9 parts, 30 seconds long, still didn't work. Is there a solution to this?
I have the picture named "02.png", I rechecked if I made a typo, yet it says there's an error
your file has to be png
You need to make sure the png/jpg is typed in either caps or lower case
It is typed in lower case
Okay it all seems cool and I'm grateful for this but the part of using specifically vegas pro and 256 res isn't a bit unnecessary? Like you could use any editing software and the configuration you prefer
tried to make one. somehow the colab page looks different and it always has the same error
def thumbnail (file):
return imageio.get_reader(file, mode=´I´, format=´FFMEPG´).get_next_data()
Help I wanna do another one, but when i try to create the deepfake the it says something like 02.png is missing something like that
you just need a png file inside your destination path
it should be named exactly the same.
i love that you actually link your music!!
for those who failed, the image must be a .png
How does one convert a jpg image into a png? This would definitely help me out
@@tobreezzyproductions you can convert jpg to png using paint
The file res must be 256x256 at least
FileNotFoundError Traceback (most recent call last)
in ()
1 from demo import load_checkpoints
2 generator, kp_detector = load_checkpoints(config_path='config/vox-256.yaml',
----> 3 checkpoint_path='/content/gdrive/My Drive/first-order-motion-model/vox-cpk.pth.tar')
3 frames
/usr/local/lib/python3.6/dist-packages/torch/serialization.py in __init__(self, name, mode)
208 class _open_file(_opener):
209 def __init__(self, name, mode):
--> 210 super(_open_file, self).__init__(open(name, mode))
211
212 def __exit__(self, *args):
FileNotFoundError: [Errno 2] No such file or directory: '/content/gdrive/My Drive/first-order-motion-model/vox-cpk.pth.tar'
@@jameslestero.rodriguez1626 looks like something didn't copy over right?
please make a version of the file on google drive. that way you dont actually have to download just import my internet is like 10kbps i cant download and then upload it somewhere else
@@salmonslay i figured it out, if you go to that collab.google site the files are in the drive link inside there, just add them to your own drive :)
im just making this meme with everything i see lmao
Where i found the dame dane file?? i dont found it in the description
bro another thing, the links in the description are buged or something because i can't download it cause there are a lot of annoying spam
I can't do it, as soon as I click on in "Load Driving Video and Source Image", I get an error message, can you please help me?
"#Resize image and video to 256x256" probably is because of that
I'm Brazilian
this video helped me a lot thanks :D
Ah yes Filipino RUclipsr with an international audience. Vemry rare.
Tru
Mostly when it's Filipino
The comments will be:
Hug 2 hug Po!
1 like 1 subscribe
Back to you.
That's so fkin annoying
@@FrixAnims I feel your pain brother.
@@HelloHello8D ikr
@@FrixAnims Normies be doing that lol
Excellent, 4 and half minutes, nothing wasted.
ModuleNotFoundError: no module named 'demo'
Im stucked on that point the Create a model and load checkpoints
same
@@Axesmed someone please
same here :/
click the first 2 checkpoints and you're good to go ;)
what checkpoints bro?
IT WORKED! Keep up the great work.
2:33 this becomes red when I try
Same
change your picture file to .png
@@kirigayakazuto843 Bro demo module is basically not found in that repository, why the fk should someone change picture file to png? when its not png they couldn't pass on the checkpoint before
change .png to jpg or PNG if that doesnt work
Same
YOUR A ABSOLUTE GOD!!! I made moist critikal singing it to me! And also guys other videos work to not only Baka Mitai.
2:43 I got an error from this step, the red play button
Yeah Some
Same
How To fix that
you need the cpk archives in your folder
@@dajologo6811 what cpk
@@dajologo6811 What archives
When I mount the google drive which is linking my google drive account it did not show the part where i need to go to the URL browser & copy the authentication code instead it shows the "Mounted at /content/drive, how can I fix this issue?
I managed to make one, but as soon as I want to make another one, I download it but it appears to be the same image I've used for the first one rather than the second image that I have put in even though I did all the steps again just to change it. But it's still persistent on using the previous image. If that makes sense?
help anyone?
i have this problem too.
having this problem too hope i can get a fix
use another google account
found the fix, go to runtime on the top and do factory reset runtime
Also if it don't work when you're loading the video to deepfake it, Replace :
driving_video = imageio.mimread('/content/gdrive/My Drive/¤¤¤¤¤¤.mp4')
by
driving_video = []
reader = imageio.get_reader('/content/gdrive/My Drive/¤¤¤¤¤¤.mp4')
try:
for im in reader:
driving_video.append(im)
except RuntimeError:
pass
It actually helped! Thank you so much bro!
ModuleNotFoundError: No module named 'demo' keep getting this error
Same, did you find any solution?
You need to download the files given in the description
and paste it on your first-order-motion-model folder
@@happypiak1391 already downloaded and uploded on my drive. The problem is. In the first 3play button, it ks working but when generating the final output, it always says modulenotfounderror
I want to make araki sing dame da ne
Making memes with the boys at 4:20 am
fatal: destination path 'first-order-model' already exists and is not an empty directory.
ruclips.net/video/7CK8wqsvOwo/видео.html
it still not working at the
from demo import load_checkpoints
generator, kp_detector = load_checkpoints(config_path='config/vox-256.yaml',
checkpoint_path='/content/gdrive/My Drive/first-order-motion-model/vox-cpk.pth.tar')
the cell play button turns red on the "import imageio" thingy how
Change 02.png by 02. the extension of your image in the python code
@@remi884 it's ok if it's jpg?
@@rubenraffs yeah
Best guide on YT for this
im having an error that says "no gpu backend"
it worked like two days ago i "might" have a fix
Just wait for 1 day (not sure) and it will be fixed automatically
lol i cant tell if your joking or not
it works now
There is a twelve hour delay until you can use it again
can i make something other with this? like someone say duck you?
I want to do this but there I’m afraid my face cannot fix while singing it.
You can always take another or use anything else, being creative is the key to content creation
Awesome tutorial music!!! Love it
me and my bois learning how to code so that we could make a meme
NOTE: For those who don't have Sony Vegas, you can use something else like Camtasia, Openshot or other video editors you have on your PC.
Even ClipChamp?
So I was able to make a couple with the same template then when I changed the video, it gave me an error at the checkpoint step, tried it every which way and eventually used the same template and image I already used (and previously worked) and now it’s not working for me at all, anyone know what to do??
me too
same prob
edit: i used incognito mode and it worked just fine
The fourth piece of code doesn't work. The button turns red and when I hover over it it says "Previous execution ended unsuccessfully", help lol
I get stuck at the "load driving video and source image" part
Yea this tutorial is not really helping
@ I FIGURED IT OUT! if you don't use the specific source stuff, it will not work. Theres another tutorial with easier links.
@@SlurpyTheDog give link pls
@@enchantedskullbit ruclips.net/video/5CgrtjZbj_8/видео.html
best tutorial dame da ne i could find😂good job bro!
what if we wanna make another one?
do we need to redo all the steps like reuploading the zip files?
no just replace the 02.png with a new one then continue iwth the rest of the steps
@@lavaking5990 i tried but it wont let me redo it and just shows a red play symbol
@@lavaking5990 pls educate
there is also a limitation for the gpu usage in collab
I have a problem. The first time it worked but when i want to try it with an other Image the third "codeblock" (idk) says: " Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount("/content/gdrive", force_remount=True). " how do i fix that?
same here
@@vinimaiscedo it means the drive already mounted,
as long as you're not disconnected, you don't have to re-run that cell or codeblock
Me who doesn't have a pc: *DAME DA NEEE DAME YO*
Yo you can do it from your phone too, on RUclips there's a video where a man explain how to do a Baka Miati from phone.
dame na no yo
Poor gang ☹︎
@@emanueleaiosa736 link
Pain.
can you use the both files for any other deepfake or only the baka mitai deep fake?
[Errno 2] No such file or directory: 'first-order-model' /content/first-order-model
Help
is picture png?
@@kade9427 yes
me 2. What's wrong?
Wait, I guess that u have to extract the files pth, but it only works in mac
Just reload the page
How do I do it again if I already did it once..?
Can this be done with Filmora?
It can be done in any video editor, just speed the clip up by 150%
@@imminent3178 when i did it it was 150%
@@ryandatta6059 my bad, I'll edit my reply. Thanks for letting me know
im always stuck at 1:32, when i click the play button the place where im supposed to put the authorization code doesnt appear
I wanna do this multiple times but it says the driver is already mounted
change the line that says drive.mount('/content/gdrive')
to drive.mount('/content/gdrive', force_remount=True)
@@butthog9663 or just skip the "Mount your Google drive folder on Colab" part
Love you man! I've got so many troubles doing this but i finally made it!
Do you know how to restart that program
@@lukeknd unfortunately not
Bruv, can't download the video . "generated.mp4" is not created im done with everything.
UPDATE:- The site might glitch it took me a couple of redos but i was able to get the "generated.mp4" in the end .
It happened the same to me but I found it in the "first-order-model" folder
Update:- i was able to download the file just had to refresh and redo the process a couple of times.
I have a question,is this safe like the colab link,do they access your account and use the picture to do bad things?
Can this be done with davinci resolve
EDIT: yes it can, also the whole colab google link cant connect to GPU-backend now
does it work with premiere pro?
@@carlabounasr3292 yes all u have to do is change the clip seed so any editing software is fine
it was all good for the first time but now i want to deepfake multiple images and i can't do multiple, only worked for the first time :/
ruclips.net/video/7CK8wqsvOwo/видео.html
someone have the files in mediafire or another page¿
just copy the vox file to your Gdrive from this link drive.google.com/drive/folders/1kZ1gCnpfU0BnpdU47pLM_TQ6RypDDqgw?usp=sharing
@Matt Andrew Sambile 👌
@@anakkampret7352 yo where's the template?
@@anakkampret7352 you're a fucking legend, ty dude
nvm, got it
Aye it worked. Thanks my guy!
All this for dame da ne
Noixe its a good tp in quarantine
Is it possible with video editing programs other than Sony Vegas ?
When I Click The Perform Image Animation I showed a red Play Button
same
same
for people who are having the FileNotFound problem and youre POSITIVE all the labels are correct, make sure the google drive account is the same account you are logged into at the time if you have multiple accounts. took me days to realize i was logged into the wrong account lol, then it worked.
Me: clicks the download link
Viruses: Allow us to introduce ourselves
Can someone give the normal link to the downloads? I have 2 viruses on my phone because od him
Is it possible to use other deepfake templates besides the Dame Da Ne template? If so, how do you resize the video template?
man can somebody like make some shit where I just put the pic in and it does all the work
can you make a google drive mirror? it's easier to just import it to google drive than downloading it and uploading it, also i have a shitty pc with no storage to spare
Okay, after trying a few things I found a way, here's the fix for generated.mp4:
See the code that says imageio.mimsave(..................)?
substitute it for:
imageio.mimsave('/content/gdrive/My Drive/first-order-motion-model/generated.mp4', [img_as_ubyte(frame) for frame in predictions])
Now, you just go to the folder in which you put the base image and video, and you will find the video there. it's a bit slow, for some reason, but you can fix it in vegas at least
You can subscribe if you want to thank me for some reason
thanks mate
@EDM Sói idk lol im about to test it out as we speak
@EDM Sói ack... there's a filenotfound error somewhere in the load of the driving video and source image
yeah, seems to be something wrong with the python code itself
@@fraddyfuzzybear2756 That happened to me too before, what I did was just redo all the steps, I don't think it's a fault of python, it just doesn't find the required stuff OR the cd changes. Do cd again (the second code segment)
You learned how to make this meme and on top of that, a free video editing software. Thanks man xD
can this be made in premiere doe???
yes speed the video up to 300%
@@MirForge so you just need to speed it up and that's it?
@@rafael.s thank you
Sweet baby jesus those are some big files.
I've been making a bunch of this and shouting you out in the description! Thanks man!
4:03 lol
you can add music to the video using ffmpeg:
ffmpeg -i bakamitai_template.mp4 -i generated.mp4 -c copy -map 0:0 -map 1:0 your_meme.mp4
at where i put this code?
ok basic question, how do i save the product as an mp4 instead of a vegas project
export it as mp4
I did it, but after making a couple videos, I got an error message when I tried to connect again saying (Cannot connect to GPU backend).
Anyone know what this means and how to fix it?
fatal: destination path 'first-order-model' already exists and is not an empty directory
if you have problems maybe the drive need to be cleaned, i tried two times and the way that the drive is empty works
:)
Help pls >>> RuntimeError: cuda runtime error (100) : no CUDA-capable device is detected at /pytorch/aten/src/THC/THCGeneral.cpp:47
btw if you cant find the generated.mp4 its under first-order-model
when i click create model and load checkpoint it goes red button n stops working
I try to click the "Load driving video and source image" block and this is the error I get:
RuntimeError: imageio.mimread() has read over 256 MiB of image data.
Stopped to avoid memory problems. Use imageio.get_reader() or memtest=False.
I have no knowledge of coding, how do I fix this?
under the play button for that block you'll see something that says 'imageio.mimread(some text in red)', replace the 'imageio.mimread' with 'imageio.get_reader' and run it
@@s4fwan Thanks so much (not that guy)