I’m interested in taking an object that uses one input UV texture and mapping and texture baking it to a new output UV mapping and image. Can anyone tell me if that is possible and how I might go about it?
What I’d like to do is to take a simple photograph of a face and UV map it to a mesh representing the face. As you might expect, it is difficult to lay out the input mapping and, due to shadows (along the underside of the chin for example) and undesirable surface orientations (along the sides of the face for example), it is essentially impossible to obtain from the photograph a complete, continuous and convenient mapping of the entire face.
So, since I’m no artist, my hope is that I can map as much useful surface area from the photograph and then texture bake it to a standard face mapping in which the complete surface of the mesh is more or less (weirdly by appearance) spread out smoothly over a flat surface. I can then filter out color discontinuities from that image and make other improvements. That should allow me to produce a nearly perfect and realistic representation of the face and apply my own lighting, shadows and such.
Any ideas? Thanks.