A simple two color selection based on B-, z-, and K- band photometry is proposed for culling galaxies at 1.4<z<2.5 in K-selected samples and classifying them as star-forming or passive systems. The method is calibrated on the highly complete spectroscopic redshift database of the K20 survey, verified with simulations and tested on other datasets. Requiring BzK=(z-K)-(B-z)>-0.2 (AB) allows to select actively star-forming galaxies at z>1.4, independently on their dust reddening. Instead, objects with BzK<-0.2 and (z-K)>2.5 (AB) colors include passively evolving galaxies at z>1.4, often with spheroidal morphologies. Simple recipes to estimate the reddening, SFRs and masses of BzK-selected galaxies are derived, and calibrated on K<20 galaxies. Based on their UV (reddening-corrected), X-ray and radio luminosities, the BzK-selected star-forming galaxies with K<20 turn out to have average SFR ~ 200 Msun yr^-1, and median reddening E(B-V)~0.4. Besides missing the passively evolving galaxies, the UV selection appears to miss some relevant fraction of the z~2 star-forming galaxies with K<20, and hence of the (obscured) star-formation rate density at this redshift. The high SFRs and masses add to other existing evidence that these z=2 star-forming galaxies may be among the precursors of z=0 early-type galaxies. Theoretical models cannot reproduce simultaneously the space density of both passively evolving and highly star-forming galaxies at z=2. In view of Spitzer Space Telescope observations, an analogous technique based on the RJL photometry is proposed to complement the BzK selection and to identify massive galaxies at 2.5<z<4.0. These color criteria should help in completing the census of the stellar mass and of the star-formation rate density at high redshift (abridged). Comment: 19 pages, 17 figures, to appear on ApJ (20 December 2004 issue)