The best debate concerning this was between Christopher Hitchens and Professor of Mathematics at the University of Oxford, John Lennox. You probably will not see a better cerebral match up about the question and validity of religion. John Lennox as a scientist and also a devout christian who gave numerous historical references in regards to what he perceived as empirical evidence. Most would assume considering his background in the hard sciences, he would be centered on logic and order for justification. This was were he based that irrefutable proof. However, that perspective is open to much attack and interpretation.
Richard Dawkins gives a great argument against - being a scientist himself and looking at the anti-order/chaos and randomness of the evolutionary creation of life. As a result, I do not see the empirical proof based on what senses we have as human beings. In essence, it becomes an easy cop out strategy to refer to the empirical by such bombastic highly self interpretative means of defining proof. To be honest, if empirical evidence was needed - one must first come to define what constitutes empirical evidence in light of the question. There must be an agreement. Hard evidence no- soft evidence - yes/no depending on your belief, so in essence like the Hitchens/Lenox debate the answers were equally convincing but depending on your perspective unchanging. In the end, if you are religious there tends to be a more"open" definition of what constitutes empirical which in my book seems highly dubious since that very "open" perspective is not all encompassing to other facets of life. ie: religious fundamentalism.