What are you there for, then?

I started doing the same with md5 hash filtering but it was very slow going, an AI that can detect low quality posts and filter them out is the holy grail I guess

Not only kpop, but entire "fandoms" as a whole. When that software finally arrives, I want to be able to refer back to this comment and say I read it here first.

I guess I had developed a kind of builtin ad filter into my brain, from ignoring so many ads over the years.

Thankfully, my place of work had reasonable people working there. But I think if I worked somewhere more strict even that one ad might have prompted a conversation with HR. Nonetheless, I promptly installed an ad-blocker on the browser in case any ads would be shown in the future that would have been more explicit than the one that had been showing this time.

Once an image tested positive with your first model, you could run a second more CPU intensive model (but also more accurate) on it.

Because I live stream and people try to trick me into viewing NSFW content on stream.

Is this really a question?

The issue with NSFW images is not you seeing it on your own computer, it is other people seeing it on your computer (e.g. while walking by or for some other reason).

Its the training that takes forever, due to the fact all those numbers need tweaking. However one you have the model, classifying is pretty fast.

So I decided to move the inference to the server.

It's fast but I didn't toy with GPU setting when running Tensorflow model on JS.

So seems this extension is not really for your use case but for others.

My favorite thing is that they had a JarJar filter for the Star Wars prequels.

Crowdsourcing such metadata would be a good start, although it can be subjective.

If you find out a better way for streaming media, please share.

Will flag sections of the video with explicit / suggestive / partial nudity

[1] https://en.m.wikipedia.org/wiki/CleanFlicks

Vid Angel took a different approach, allowing you to fast forward over parts, and appears to still be operating: https://www.vidangel.com/

They had an ever expanding catalog of films they had processed to avoid any nudity or bad language. I think they cut those parts out entirety.

Here I found it.

They got sued out of existence apparently by the movies studios but was recreated and they now somehow do it with streaming content from Netflix or HBO.

I dont know how they do that but I figure they will be sued again.

I am not endorsing the company. I have never used it and never will but it is out there.

https://www.vidangel.com/

Surely a much easier thing to do would be to simply stop being offended by the human body? Just watch the movies anyway. Your kids won't explode if they see a boob.

This reminds me of when I went thru all different low calorie sweeteners to put in coffee until I realized this is a problem I created myself and coffee is actually better if I just get used to having it plain.

There’s a whole genre of family friendly films, those will serve you better than trying to hack up stories for full adults.

The answer is "American Prudery".

One film I think of in this regard is Schindler's List. I've seen both the original and the TV edit that Spielberg made to broadcast for teenagers to see. The primary edits I noticed were sex scenes and post-sex nudity. Those scenes weren't gratuitous(they showed something about the characters) or anything, but were edited(either cut or blacked out part of the frame). BUT there was still lots of nudity in the edited film. This was nudity of a different nature. People forced to strip down and paraded in the open for inspection.

Sometimes parents just want to take the easy path and consider all nudity as out of bounds for their kids. I think that's fine.

Actually I think a lot of people find it awkward to watch graphic sex scenes and extended nudity with their family. As a grown up, the I feel awkward watching parts of game of thrones with my parents.

That's a shame because I would like to watch this show with them. Lately a lot of major TV shows seem to aim to include one nude scene per episode - I guess the money people think that's what sells...

I can understand why some people might find violence equally difficult, but where it features it tends to be much more tied to the plot.

Some of my CX team is getting harassed by trolls who send profane messages and racial slurs through the Zendesk Chat interface. Zendesk does not have a way to filter out profanity (for Chat at least; there are options for emails or tickets).

I’ve found simple client side extensions that can censor words, but a better approach would use NLP to grasp the context.

“I hope you die” contains the same words as “My <product> got wet, I hope it didn’t die”, but with vastly different intention.

Sex is part of human nature – and, I would argue, a much bigger and better part of it than violence and aggression. Why do we hide from it?