I Traced Homer 500 Times 

yw zerkaa's brother

W video you deserve it

111k😤

crazy?? i was cra--

160k now

Mmm... Bacon...

the pfp mix-

That's perfect lmao

It always get more black.

it's like watching him deflate like a baloon lmao

Real

Truth

No but ok.

@@gillhewerfamily6550 bro it was joke

@@Dani4You-mw9md🤪🤪🤪🤪🤪

Bruh that was a cheap dig at something that isnt even true

Reminds me of the simpson’s travel into the future couch gag.

@@giggitymaster8799yes it is, the Simpsons has declined severely over the years

@@dinnerboons1504 FAMILY MEET ME AT THE KITCHEN CUUUUUUBE

@@giggitymaster8799ya the quality of these drawings are far better still.

Stop- i can imagine marge saying that

Oh no

I read that in Marge's voice 🤣

but MARRRGE

​​@@Matsonius"I like continuity errors"

hehe

Why you little!

Nice pun, ol' man.

HEY!!!! LOL!!!

His ear progression 📉📉📉📉

His progression 📈📈📉📈📉📉📉📈📉📈📈📉📉📈

megamind

LOL

He went from Simpsons to Code Lyoko

Stylistically though, animators who are wishing to put more than 3 minutes into each frame using this method can create some fantastic end results. Some verry trippy sequences have been accomplished by those who harnessed the potential of "straight ahead" instead of pose to pose.

​@@PileOfStones got some examples in mind?

@orangy57 Try telling that to the crew who animated Steven Universe.

​@@MultiPBStationArchivethat doesn't really apply here, what was inconsistent in steven universe was character models from episode to episode, which was intentional

thin brain

hes not smrt anymore

more like a strangely inflated tumored one

more like a strangely inflated tumored one

Homer becoming idiot​@@MightyEFX

M too

Something something Matt Groening Signature

By the way, did you notice how the ear slid down his face?

It's interesting how all the shapes still existed for the most part throughout, except the right collar (to us) which got shredded very quickly

@@bobogus7559 I did

Oh yes, a refference we can all relate to

reverse mewing

relatable

Same

lol

That’s a little too relatable…

Into a bush

@@NintendoBoy64 and got mauled by several raccoons

Wouldn’t it be Homelessr?

@@GabrielPerez-br5mm yeah, maybe idk lmao

The living line!

Sort of; they don't trace over the previous lines every frame they're idle, they just trace over it once or twice and cycle through those frames as long as they're idle. Source: Huge EENE fan.

Father redhead???

@@porcelainghoul WHY DID IT DO THAT?! I meant forehead lol

This was probably my favorite part. Starts out normal and then ends up at like chin level.

😂😂😂

I turned smoothing down to 90 for my next thing and it's a life changer

It's the same hand drawing these

"I am Simpson!"

*ALL HAIL THE DARK LORD OF THE TWIN MOOOOONS*

Okay and?

Well, Womp Womp Wompity WOMP.

*Womp Fucking Womp*

pmoW pmoW

Womp fuckin womp

Fr

what happened in the last week to get this an additonal 85K+ views lol

How does it feel eating uranium

I need to know what that is.

​@@mikebliss3153Check the description.

The drawing is like dementia.

My favourite sister

Top two favourite brother

Speaking of AI, you know how diffusion models work by like, being trained to remove noise added to an image, at a pixel level? I wonder, if we used some kind of vector graphics type thing, if we could introduce noise/errors in a way that emulates this kind of tracing, Could one train a model to do it in reverse? Might be interesting.

​@@drdca8263 bro is yapping

@@melonga7435 I was just thinking that it might be a neat way to like- well, if someone hasn’t yet gained the skills to draw curves and lines that look nice (having them be too shaky and such), it might be cool if an ML model could try to estimate what the error was, as a tool to help a novice produce better line work. In machine learning, in order to produce a large dataset, something which is often useful is to take a process which we know how to easily do automatically, but don’t have a good process for un-doing it yet, and then do the forward process on many inputs, And train the model to, given the output, try to produce the original input. And, it seems to me like the process of “take these curves, and then trace over them, but with some error”, would probably be something not too hard to do automatically (though, not in a way that quite captures all the details of the ways a human tracing would actually tend to differ from the original. But possibly close enough?) So, if one had a dataset of line drawings (perhaps one could obtain this by applying edge detection+smoothing to a collection of public domain works (maybe photographs) ), one could maybe use this process to train a model to take sloppy line-drawings and produce less-sloppy line-drawings. Really, this would just be for an improvement of the line smoothing features in photoshop, and gimp, and other similar digital art programs. (Only, it would apply, I suppose, to a full image, rather than to a single curve?) Edit: that being said, I can also imagine such a tool being used as a crutch, in a way that could be detrimental to one gaining certain skills, so it might not be great to use. Idk. Just an idea.

Should be an app where people trace each others picture, from original to end result, in certain amount steps, like "telephone". And then the app creates the animation, maybe backwards so it would be more of a surprise what original image is. Would be fun

​@@DaP84gartic phone is that game

@@sain1536 okay cool!😀

Go for it

@@CyanSandwich Yay!

Ima also go for it

how did you turn tracing homer 500 times into a poetic masterpiece 😭

The original thumbnail had that, but this one seemed to do better