One visualization strategy is to conceptualize the figure as a mannikin rather than become too enamored of the anatomical details. A generalized approximation of geometry that resembles the masses of the figure emulates how the sculptor builds the volumes. For example, a peanut shape could be analogous with the torso, an ovoid for the head, cylinders for the arms and legs, etc.These primitive geometric forms are more easily mapped in the brain as we shall later find out, and create a volumetric placeholder which is ultimately more accurate than a finely placed contour.
What we are trying to train ourselves to see is the complete form in all its volumetric occupancy.
Illustrated here are CAD renderings of the human figure as groups of polygons that are "skinned" over a volume. From these renderings, it's possible to see more clearly the inner walls of the reverse side of the figure. Also noticeable is how the inner reverse wall exhibits the odd perceptual phenomena of appearing both concave and convex at the same time. Seeing through a wireframe CAD drawing or cross-sectioned rendering allows one to see the bi-symmetrically of the volume very clearly. In this example I sectioned a human figure into two halves along a natural lateral curve, dividing the figure into anterior and posterior halves. Separating the two halves it is possible to see the shell of the hidden back half of the volume.
The last picture illustrates the bi-lateral symmetry in action. I have duplicated the front half and mirrored its symmetry. By comparing the inner shell with the outer shell, it is virtually impossible to tell if the figure is convex or concave.
Convex or Concave???
