Yeah, outlines like subway lines where each colour is visible the entire way would ne nice. Might be a bit thick in some places though.
Yeah, outlines like subway lines where each colour is visible the entire way would ne nice. Might be a bit thick in some places though.
Why not train a model on exactly one work (effectively encoding the work in the model)? Is the ability to produce something else an argument against copyright infringement?
If the definition of AI is broad enough, train a video encoder and make the act of viewing a video file copyright stripping!