I think so, depending on how you intrepret the objects and whether you ever see the back sides of them.
Interpretation one: Each object is a hemisphere with a flat surface partially cut away. You could make this pretty easily with a custom bevel and a transparency map.
Interpretation two: Each object is a cylinder with a flat surface cut partially away as well as the side of the cylinder partially cut away, the cylinders are angled to give the appearance you see. Again, pretty easy to do with transparency maps.
In both cases, the illusion falls apart when you rotate the objects to reveal the sides of the model not surfaced.
Which interpretation are you going for - or is there another that I'm missing?
Thanks for chiming in. Essentially, I have to use these two parts to create a transition.
Visualize this -
The two parts come in, off screen, from their respective upper left and lower right corners and converge in the middle. They rotate a little so that when they converge for about a frame or two the cover the entire screen - never mind if the outer edges get cutoff.
They then move away form each other to end where they are now.
So, yeah, there's got to be some rotation to allow the bevelled parts to match up to each other.