I believe you're just referring to transforming each marginal distribution to $U[0,1]$ via the [probability integral transform](http://en.wikipedia.org/wiki/Probability_integral_transform), which when applied to each of the variables individually, transforms a d-dimensional distribution to its copula.

For example, if you had a bivariate normal $(X,Y)$, and transform $U=F_X^{-1}(X)$ and $V=F_Y^{-1}(Y)$, then $(U,V)$ is a Gaussian copula.

e.g. see [here](http://en.wikipedia.org/wiki/Copula_%28probability_theory%29#Mathematical_definition)


There are some recommended introductory readings [here](http://stats.stackexchange.com/questions/37951/introductory-reading-on-copulas?rq=1)