We've seen one or two really good-looking examples posted here, but a lot of people still share my opinion that the photo-mapping doesn't work very well yet. Depends on the face you're doing, I suppose: some faces fall within Don or Judy's morph range, some don't. Then there's this whole other issue about why the resulting heads are so often shrunken. (But you can scale them back up.)

It does seem to be working better since the patch, which makes its control mechanism more visible. Also, I started getting much better results after applying a suggestion I saw somewhere here in the forum (which I should have thanked and credited at the time, when I still remembered who posted it):
For the full-face shot, don't use the picture you snapped, but instead go into your graphics program and create an absolutely symmetrical picture by flipping half of your face over the other. You'll have to experiment with both sides and exact placement before you get something that still looks like you (an absolutely neutral expression is vital), but this kind of bi-lateral symmetry is what the Face Room wants to see; the control points are much more likely to fall into place and give you a likeness-shape from such a starting point.

I repeat, though, I'm counting on CL to keep improving this feature: it's a great start, but it's not nearly there yet.