Consumer 3D LCD monitors are becoming much more readily available and cheaper, and often come packaged with 3D glasses. My understanding is that many of these displays are passive, with every second row of pixels polarised in one way, and every other row polarised differently. With the glasses on, your left eye sees one set of rows and the right eye sees the other. Thus with a suitable image sent to the monitor, and for those people fortunate to have binocular vision, you can see a 3D image.
I am interested in experimenting with this technology by controlling the two separate images myself. Specifically, given the usual source image that would be sent to the monitor, I would like to transform (e.g. translate & rotate) every second row of pixels relative to every other row of pixels, as demonstrated in the image.
I'd like to be able to control the amount of translation, rotation, etc., but have this relatively independent from the normal functioning of the computer (ideally in a Windows environment, e.g. Win7). I imagine this could be achieved at the device driver level, but while I've had a fair amount of programming experience, I have none when it comes to the driver/hardware level.
I am wondering what methods I could investigate in order to achieve this? My overall goal is to be able to use a 3D LCD screen (or similar), but have complete control of the relative positioning of the (2D) left & right images. Seeing as 3D LCD monitors are relatively cheap nowadays, the even/odd row of pixels seems like a possible way to go (I am not concerned with the decreased resolution of this approach). I could certainly write my own standalone software to perform the pixel-by-pixel transformations, but this would be limited to whatever images are shown in the software itself, and would not be compatible with watching TV shows, movies, games, etc. on the computer. Any advice or suggestions you may have on customizing/programming 3D LCD displays would be highly appreciated!