With training, a tennis player can see where the ball coming towards her will strike the ground. Why not say this is genuine perception? That it takes training is irrelevant—most perceptions do. I guess the best reason to deny this is a perception is to require that a perception be caused by the perceived state of affairs. But now here is an interesting thing. This morning I saw a cow and I saw it as a cow. The causal criterion requires, then, that the cowness of the cow have caused my perception. And this requires an Aristotelian view of forms as causally efficacious. A Thus, it seems, either we can see the future or Aristelianism is true or we can't see any cow to be a cow. Maybe the causal criterion can be replaced with an explanatory one, though, which would weaken the Aristotelian conclusion while broadening the scope of what we can see.
[By the way, sorry for the typos in recent posts. They were typed mainly with vim over ssh on my Treo. I now have a Internet access on a device with a full-size keyboard, which should improve things.]