Decensoring Hentai with Deep Neural Networks
github.com
github.com
https://arxiv.org/pdf/1806.03589.pdf
http://jiahuiyu.com/deepfill2/
I think the most interesting part of this project would be generating the dataset. That is a lot of manual redrawing of vaginas/penises. Or the reverse, where you find already decensored stuff and add realistic censorship yourself. Either way, a lot of time will be spent tagging genitals to build a large enough dataset to make NN techniques work...
My intuition is that the dataset will be too imbalanced to learn anything useful. Even if you crawl only decensored images, the area you truly care about inpainting is still pretty small. If you don't focus on it somehow, it might learn how to inpaint anime-style geometry correctly (from the rest of the image) but produce "barbie doll" style anatomy.
So I'm guessing that this model doesn't have any specific training for filling in only genitals. I think there are category-aware inpainting algorithms that could be used, but they'd need tagged data, as you say.
Presumably if you trained a really big inpainting CNN a lot, it would learn genitals (along with everything else), but it's understandable that he would try a much more targeted approach.
So you just need to find both versions in order to make your dataset. Add the large amount of decensored artwork and it is not that hard to build a huge dataset. It may even be automatable by parsing imageboards.
Phase I: Deep learning combined with realistic physics simulation will eventually replace human actors. At some point amateurs can record their movements, then use variational autoencoders to improve their features (bigger boobs, thinner waist, smooth skin) and replace professionals. This is 'Uber phase' for porn and drives down the wages.
Phase II: Eventually AI can mimic complete scenes and adjust the 'plot' to the current state of the observer. Tight sexual biofeedback loop will form.
Phase III: The rapture of the nerds. Fully interactive VR + toolset. In the context of the Fermi paradox, Phase III is the Great Filter.
----
It's way easier to remove call centers than coming up with physics simulations that replace humans. Also I think porn is incredibly cheap to produce so everything must be automated to beat the current production cost.
A market for verifiable human pornography will exist alongside the new porn.
Also, have we already dubbed virtual ML pornography "deep throat"?
No, that term is already established in the industry and means something else entirely.
This is why the ML pornography excites me. All my twisted fetishes that nobody caters to can finally be fulfilled.
Edit: should have mentioned NSFW
You'd need I, be it AI or the normal kind, to invent a guess from a single frame. But watch frames that each differ a little over time, and you should be able to deobfuscate pixellation and fuzz with plain math.
Conceptually simple. Not saying I could knock it out over the weekend, just that there doesn't seem to be a mystery about how to proceed that puts it into "throw some ai at it and see if anything happens by ai magic" territory.
I'm sure it'll be a useful tool with refinement, but this is still a solution to an artificial problem. There's no reason for Japan's current censor laws to exist in 2018.
Judging by the examples, correctly-coloured yet featureless skin.
Where it might be a little more clear cut is feeding the market that clearly exists for “teen” porn, which is currently filled by 18+ actresses. I suspect a non-trivial percentage of the non-child-abusing population would consume simulated 16 year old porn without ever considering the real thing. In that case you might see some changes.
Reply edit It’s already outlawed. In the US for example pornographic drawings of children are outlawed. In the same way it’s illegal to try and sell cocaine stimulants, it’s illegal to try and sell child pornography simulations. I doubt the A.I. itself would be illegal, just its output.
PS: I am obviously against child pornography, I'm just curious about the implications of being able to generate illegal content and how governments would deal with it.
However, a neural network represents a mathematical model, not a specific "real world event". Although I suppose you might argue that since the dataset required to generate it would be illegal, it is illegal?
I think linking illegal depictions to illegal events would be saner, but that's not how it currently works.
Come to think of it, a couple of years ago there was this app called FaceApp which came out which used neural networks to modify selfies in funny ways (like making you older/younger/of the opposite sex).
I wonder if anyone has ever run a pornographic image with the "younger" filter in the app. Would that be illegal?
This is sovereign-citizen thinking. It's like trying to crash people's phones by yelling "Hey Google, this sentence is false!" at them.
An even cheekier way to drive the point home would be for a sex-stories website to have a slider in the footer to update all age values across the site.
For law enforcement purposes, I think it would be similar to a printing press: owning one is not a problem, but inserting printing rolls that look like dollar bills most definitely is.
https://en.wikipedia.org/wiki/Ashcroft_v._Free_Speech_Coalit...