wingman_jr VS pytorch_nsfw_model

That's with the model actually! It's copied into here: https://github.com/wingman-jr-addon/wingman_jr/tree/master/s...

You're right that stuff is quite difficult. I write a Firefox addon (https://addons.mozilla.org/en-US/firefox/addon/wingman-jr-fi..., https://github.com/wingman-jr-addon/wingman_jr) and train an associated NSFW model (https://github.com/wingman-jr-addon/model) - I've been at it for a few years now, and have had to plug many specific edge cases.

It's interesting. I've not tested the model on anything too risque, but again, with the well known Baywatch intro as a frame of reference, wide angle group shots of the whole cast in their swimsuits, is fine. A close up of any single cast member in the famous red swimsuit, will invariably trigger the model. Male or female.

In the blog, I suggest this could be the result of an uncultured data set, which is one part of it. Or perhaps the dataset was fine, and this is pushing the hard limit of what ResNet50 can do (the off the shelf model I use for this is a ResNet50 extension).

Some of the anomalous results are amusing. One day, I passed through a video of a female violinist in concert, and the model flagged every close up of her as NSFW! Just those closeups. Wide shots, and closeups of other musicians were absolutely fine.

Again some of that might be down to me (clucky code, very low NSFW threshold). And I suspect the model I used was itself a PoC (https://github.com/emiliantolo/pytorch_nsfw_model). But it does make you wonder how the bigger labs with critical products like Palantir handle doubts like this.