tvirosi

I agree, they're awful. But on the other hand the horror market exists so somehow people enjoy seeing things like this (in certain moods).

I admit that I never really understood this speech. Could you explain why it was significant to you and what it means?

Agree. The gifs are so distracting, I could barely get through it.

Or the 47k no fly number is just a lie

Where am I supposed to look in that pdf to understand that it isn't correct or credible? It is certainly true that the model has differentiable and thus optimizable outputs.

It might have to do with the output possibly being a probability vector as opposed to a binary hash. The whole thing is thus differentiable and optimizable (if a dog image was incorrectly placed in the bad hash bucket it might only be on the border of it while the real CP corresponding to the hash is found at the probabilistic maxima of the hash bucket). Just guessing.

Probably the model weight files

I've tried inference on the python version of onnx and it usually varies between hitting a OOM limit (while with TF it works fine) to being an order of magnitude slower. Even if the codebase is still being changed I don't see much reason for people to use it other than as a convenient distribution format.

Makes me wonder if there's a possibility of e.g. faces on fbi's most wanted being snuck into the dataset somewhere in the chain.

Or for criminals to generate perceptually similar illegal images that are no longer triggered as a 'bad' hash.

Really disgusting idea: I wonder if it's possible for someone to use this as a 'discriminator' in a GAN to configure a generator to recreate the CP this is trying to avoid distributing in the first place.

This might totally work and it's kind of impressive if it does. I'm still biased towards ultra skepticism towards all of this since the trustworthiness of all demos like this is completely corrupted at this point due to cherry picking and other deceptive tricks.