Face Shape and Reflectance Acquisition using a Multispectral Light Stage

In this thesis, we discuss the design and calibration (geometric and radiometric) of a novel shape and reflectance acquisition device called the “Multispectral Light Stage”. This device can capture highly detailed facial geometry (down to the level of skin pores detail) and Multispectral reflectance map which can be used to estimate biophysical skin parameters such as the distribution of pigmentation and blood beneath the surface of the skin. We extend the analysis of the original spherical gradient photometric stereo method to study the effects of deformed diffuse lobes on the quality of recovered surface normals. Based on our modified radiance equations, we develop a minimal image set method to recover high quality photometric normals using only four, instead of six, spherical gradient images. Using the same radiance equations, we explore a Quadratic Programming (QP) based algorithm for correction of surface normals obtained using spherical gradient photometric stereo. Based on the proposed minimal image sets method, we present a performance capture sequence that significantly reduces the data capture requirement and post-processing computational cost of existing photometric stereo based performance geometry capture methods. Furthermore, we explore the use of images captured in our Light Stage to generate stimuli images for a psychology experiment exploring the neural representation of 3D shape and texture of a human face.