Humans are able to recognize these as optical illusions. (otherwise we wouldn't be calling them "illusions")
Further, none of the optical illusions shown are so wildly "off" and "crazy" as the algorithmic goofs demonstrated in the paper referenced above.
And 0.10 popped into my head as well. However, the fact that this was even asked as a question would cause someone to stop and think a bit more analytically I would think.
