The paper presents a new approach for shape recovery based on integrating g
eometric and photometric information. We consider 3D bilaterally symmetric
objects, that is, objects which are symmetric with respect to a plane (e.g.
, faces), and their reconstruction from a single image. Both the viewpoint
and the illumination are not necessarily frontal. Furthermore, no correspon
dence between symmetric points is required.
The basic idea is that an image taken from a general, non frontal viewpoint
, under non-frontal illumination can be regarded as a pair of images. Each
image of the pair is one half of the object, taken from different viewing p
ositions and with different lighting directions. Thus, one-image-variants o
f geometric stereo and of photometric stereo can be used. Unlike the separa
te invocation of these approaches, which require point correspondence betwe
en the two images, we show that integrating the photometric and geometric i
nformation suffice to yield a dense correspondence between pairs of symmetr
ic points, and as a result, a dense shape recovery of the object. Furthermo
re, the unknown lighting and viewing parameters, are also recovered in this
process.
Unknown distant point light source, Lambertian surfaces, unknown constant a
lbedo, and weak perspective projection are assumed. The method has been imp
lemented and tested experimentally on simulated and real data.