Is it possible to reverse the polarity of already polarized light using passive materials? For a gizmo idea that I have, I want to understand if it's possible to "reverse/turn" the polarity of light using passive, relatively thin  and cheap films, similar to what we see in 3d glasses or TN screens?
By reverse/turn I mean reversing the handedness of circularly polarized light or changing the angle of linearly polarized light by 90 degrees. 
It appears that polymer or quartz rotators are available, for example http://www.u-optic.com/gid_52_prdid_62_pid_266_l_1 , but I'm unable to find a relatively cheap source. 
 A: To add to John Rennie's answer and Noldor130884's answer: a half wavelength waveplate will reverse the handedness of circularly polarized light. It's the same principle as converting linear to circularly polarized light although here you don't have the precise alignment problems (to convert linear to circularly polarized, you need to align the input light's plane of polarization at an angle of 45 degrees relative to the extraordinary/ ordinary axes of the birefringent crystal). 
An interesting aside is that, through conservation of angular momentum, the light exerts a torque on the waveplate as it does this (the classic 1936 Beth experiment actually measured this[1]).
A disadvantage of waveplates is that they are wavelength sensitive: the aimed-for conversion is set by the thickness of the plate in waves at the particular waveplate in question. So a quarter wave waveplate at 700nm will be a half wave waveplate at 350nm wavelength.
Quartz is the commonest material that will do this and halfwave waveplates are very thin - less than a millimeter in thickness. You can probably find an optically active polymer to do this. Waveplates are expensive because not many of them are made. If a consumer electronics application with a large market for waveplates were found, you can guarantee that the same waveplates would shrink radically in cost as companies engineered ways to cheaply mass produce the item in question.
[1]. Richard A. Beth, "Mechanical Detection and Measurement of the Angular Momentum of Light", Phys. Rev., 50, #2, pp115-125 1936
A: As you probably already know, reflective surfaces invert the handedness of a C-pol light. So one solution would basically put a source of light in front of you, and the other one at your back, so that it would be reflected more or less from the direction from which the first source radiates.
About thin films, I could possibly think of something, but it would be rarely something that you can use for "gizmo ideas" of yours: metamaterials. You can use a thin film as a half-wave plate (see link above), but the construction of such optics requires a lot of engineering.
