nodejs-vision VS open_nsfw

Google also used all of this to improve their OCR algorithms, almost certainly used in Google Cloud Vision[0], but I doubt this was a consideration when deciding if it was transformative/fair use.

0: https://cloud.google.com/vision

You are looking for a computer vision object recognition api, and there are several, all of which cost money. Here are a couple Google’s Microsoft’s

For an example, Google Cloud API which can read images even facial expression and much more and also can differentiate which Image is violating google terms and condition like - pronographic image

tool/software = https://cloud.google.com/vision

I think you’re underestimating how good OCR tech is. Yeah, shitty penmanship will be harder for software to interpret just as it is for the human eye, but legible ink isn’t much of a problem. I just tried scanning my number off my discs and my iPhone digitized it without issue. This technology is essentially made for digitizing handwritten text. Try it yourself if you’d like: https://cloud.google.com/vision

you guys can use this website to extract the japanese texts from the manga pages : https://cloud.google.com/vision/ , then you can google-translate the texts .

they could always add something like this https://github.com/yahoo/open_nsfw

Besides, I don't hear any solution you are providing to solve the problem of unsolicited dick pics, which is largely solved by AI at this point. The first big one I'm aware of is opennsfw, but newer stuff is also going to be available.

AFAIK Discord's NSFW filter is not a perceptual hash nor uses the NCMEC database (although that might indeed be in the pipeline elsewhere) but instead uses a ML classifier (I'm certain it doesn't use perceptual hashes as Discord doesn't have a catalogue of NSFW image hashes to compare against). I've guessed it's either open_nsfw[0] or Google's Cloud Vision since the rest of Discord's infrastructure uses Google Cloud VMs. There's a web demo available of this api[1].

0: https://github.com/yahoo/open_nsfw

1: https://cloud.google.com/vision#section-2

I suspect Apple would use a simpler model like Yahoo’s Open NSFW Model

Here's an implementation using machine learning. But in general, this is a very difficult thing to do reliably.

For adult content detection, yahoo has published a model and code. It's been some time since I tested it, but it was quite usable for pictures.