Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

This Sneaker Does Not Exist

thissneakerdoesnotexist.com

Post reply on HN
Print view View original

31–40 of 72 posts

Re: This Sneaker Does Not Exist

#31

Post by pc86 »

jstx1 wrote: post #28
fxtentacle wrote: post #26

Earlier quoted context omitted.

In most cases, GAN produces a random mixture of features extracted from real photos. That's also why you can tempt GitHub Copilot to regurgitate Carmack's swearing comment.

The usefulness of any of this generated stuff eludes me.

It depends on your definition of "useful." If you look at a single GAN-generated image in isolation it's almost certainly "useless" for most definitions. But if you need a bunch of faces, sneakers, buildings, whatever, and you want to be cognizant of people's privacy, and you don't mind the occasional weird image artifact or completely fake-looking image, these can be good alternatives to sourcing a large number of real images.

Re: This Sneaker Does Not Exist

#32

Post by the8472 »

5350-uiop-1130 wrote: post #15

maybe i'm oversimplifying. i had the same thought about the "thispersondoesnotexist" site...but are these not just composites of things that do exist? or maybe that's the point.

They are "just composites" in the sense that the NN learns to decompose an image into multidimensional concepts both local and global (the latent space) and then assemble new images from anywhere in that space even at points that weren't in the training set. Of course if the training set is small then it'll overfit to mostly reproduce the inputs.

https://openai.com/blog/multimodal-neurons/

Re: This Sneaker Does Not Exist

#33

Post by t_von_doom »

StanvdVossen wrote: post #21
injidup wrote: post #11

Earlier quoted context omitted.

Too many of those shoes look like they have a munged Nike logo which means that any company trying to ship them would soon be receiving a kindly call from the Nike megacorp lawyers. You would have to improve the training to avoid generating trademark / logo infringing designs. Could this be trivially done?

In general this would not be easy. One way this can be accomplished is by removing all shoes with such logos from the training data. But Nike and Adidas are disproportionally large parts of my training set, so this would not be feasible. The other option would be to train a machine learning model to recognize said logo's and to use this model to remove sneakers with logo's from my generated images. This could however…

Slightly different implementation of your second option: Remove the logos in the dataset and train on this brand agnostic library

Re: This Sneaker Does Not Exist

#34

Post by t_von_doom »

t_von_doom wrote: post #33
StanvdVossen wrote: post #21

Earlier quoted context omitted.

In general this would not be easy. One way this can be accomplished is by removing all shoes with such logos from the training data. But Nike and Adidas are disproportionally large parts of my training set, so this would not be feasible. The other option would be to train a machine learning model to recognize said logo's and to use this model to remove sneakers with logo's from my generated images. This could however…

Slightly different implementation of your second option: Remove the logos in the dataset and train on this brand agnostic library

EDIT: To clarify - don't remove the entire image of a sneaker, just impute away the logo filling in the space with the 'context fill' algorithm found in photo editing tools

Re: This Sneaker Does Not Exist

#35

Post by 5350-uiop-1130 »

the8472 wrote: post #32
5350-uiop-1130 wrote: post #15

maybe i'm oversimplifying. i had the same thought about the "thispersondoesnotexist" site...but are these not just composites of things that do exist? or maybe that's the point.

They are "just composites" in the sense that the NN learns to decompose an image into multidimensional concepts both local and global (the latent space) and then assemble new images from anywhere in that space even at points that weren't in the training set. Of course if the training set is small then it'll overfit to mostly reproduce the inputs. https://openai.com/blog/multimodal-neurons/

it sounds like jargon to me as idk much about neural networks. i guess its kind of a highly complex and nuanced composition. but essentially its still just a sophisticated combination of existing things.

i guess this is the same as human creativity in the sense it depends on prior experience. its more the "does not exist" that bugs me a bit.

tx for the detailed response tho, wish i was smart enough to understand.

Re: This Sneaker Does Not Exist

#36

Post by the8472 »

5350-uiop-1130 wrote: post #35
the8472 wrote: post #32

Earlier quoted context omitted.

They are "just composites" in the sense that the NN learns to decompose an image into multidimensional concepts both local and global (the latent space) and then assemble new images from anywhere in that space even at points that weren't in the training set. Of course if the training set is small then it'll overfit to mostly reproduce the inputs. https://openai.com/blog/multimodal-neurons/

it sounds like jargon to me as idk much about neural networks. i guess its kind of a highly complex and nuanced composition. but essentially its still just a sophisticated combination of existing things. i guess this is the same as human creativity in the sense it depends on prior experience. its more the "does not exist" that bugs me a bit. tx for the detailed response tho, wish i was smart enough to understand.

> i guess this is the same as human creativity in the sense it depends on prior experience.

That's roughly the idea. Although I didn't want to make that analogy because someone would try to overextend it and point out that an artist can do X and a neural network can't and therefore it's wrong.

> its more the "does not exist" that bugs me a bit.

Well, if someone learns what a sneaker generally is and then draws a fantasy-sneaker that is still recognizable as a sneaker because it stays in the general confines what is considered "a sneaker" then does that sneaker suddenly exist?

Re: This Sneaker Does Not Exist

#37

Post by angarg12 »

I work with a shoemaker and I've commissioned a few bespoke sneakers (trainers as we call them here) in the past. He comes up with new designs from scratch, and I made suggestions and ask for changes based on my taste.

Usually I browse online shops for inspiration, but I'm curious about whether this service can produce original designs sensible enough to, at least, be the base of a new shoe.

Re: This Sneaker Does Not Exist

#39

Post by got2surf »

StanvdVossen wrote: post #29
piva00 wrote: post #22

Earlier quoted context omitted.

As a bit of a sneaker head, it's been quite fun to see how the algorithm changes the shapes when playing with the sliders on the Sneaker Editor. Impressive that it isn't copying/pasting parts of shoes but actually building them pixel-by-pixel, I was editing a very AF1-looking sneaker and playing with the sliders made actual sensible changes to the shape and design. Pretty entertaining :) Quick edit: as a side questio…

Thanks! The training set included pretty much all sneakers I could find online. However, the algorithm tends to bias towards patterns that are more common e.g. hundreds of colorways of Air Jordans. BUT I can fine tune the model by training on a small subset of sneakers. So if you have any types of brands or sneakers that you would like to see generated, please feel free to mail me some examples at stan@thissneakerdoe…

It could also be interesting to define multiple subsets of sneakers - "2021 avant garde sneakers", "skate-inspired shoes", "2010s most popular shoes in the NBA" (based on https://ballershoesdb.com/ or a similar database). It could be cool to start with a model trained on a subset of sneakers, and tweak/combine things from there.
Post reply on HN