Anatomical distortions

Some papers discussing image projection from a 3D body to a shroud draped over the body to account for image distortions:

Though this image appears to be anatomically reasonable, there are some second-order characteristics which indicate that this may not be so: broadening of hips, elongation of arms and fingers, lateral displacement of hair away from side of face.

The general layout of the Shroud image suggests that it was formed by the Shroud draping over a body: this, it is natural to propose that the image distortions might be due to cloth drape. In other words, we are suggesting possible image distortions caused by projecting a three dimensional object onto a curved cloth drape, then observing the cloth flattened onto a two dimensional surface.

We note that the hips, forearm, and fingers (if they had been extended) would show positive distortion as indicated by the data. We further note that other distortions such as broadening of the hair can be observed in the Shroud image.

On the shroud, the hair seems to hand abnormally off to each side of the face, but what could have happened is that the hair was actually in contact with the sides of the face but displaced away by geometric distortion.

Examination of the Turin Shroud for image distortions
W. Ercoline, R. D. Jr., and J. Jackson
IEEE 1982 Proceedings of the International
Conference on Cybernetics and Society … %20OCR.pdf

If a cloth is appropriately laid on the front part of a body,
and a body image forms by a vertical projection on
the cloth, no major image distortions occur. Small
image distortions are to be expected, and indeed we
can observe some on the Shroud.

Notice that the sheet almost follows the fingers of
the right hand. If the sheet were flattened – and
the fingers made a vertical projective image on the
sheet – the image of the fingers would have almost
the same length as if they were straight. The appar-
ent distortion is a lengthening of the fingers. We
can verify this fact on the photograph: the mid-
dle finger apparent length is 6.1cm: measured from
(1048, 1031) to (1213, 1245) (line labeled C), that is
270px. This spans 5.9 diagonal grids, which gives
5.9 × 1.518 = 8.9cm. The middle finger of the right
hand is 10cm, as measured directly on the model. So
there is an apparent length increase of this finger by
about 2.8cm.

Notice also that if the fingers of the left hand had
been closer to the edge of the thigh – for example,
by lifting the shoulders up – an apparent length-
ening, compared to the real fingers, would have oc-
curred since a projection of the straight fingers on
the slanted sheet would have happened. This would
have been quite similar to the apparent long extended
fingers seen on the Shroud.

It was shown that, once the blood stains formed by
contact, the top half of the Shroud could not have
been lifted up to be flattened to avoid major
image distortions. After the blood stains formed, the
Shroud essentially stayed in the same position prior
to the image formations.

It was also shown that a flattening of the top half
of the Shroud is not required to avoid major image
distortions. That is, there exists some natural way
for the Shroud to lay on the body while the images
are formed without causing major image distortions
– albeit the Shroud had to be laid carefully over the
body. Moreover, it appears that there are some small
image distortions, coherent with the Shroud laying on
a human body form. … ges_formed