A spectrally based framework for realistic image synthesis

A wavelet-based framework for acquired radiometric quantity representation and accurate physical rendering

“…We assume only non dispersive objects, thus allowing a vector representation of the spectral interaction between the light and the surface. Dispersive objects should be handled by a specific data structure such as the one proposed by Sun [67]. First, because we want to test our algorithms with real measurements that are currently very difficult to obtain for BTDF.…”

“…Spectral representations rely on polynomials [55], basis functions [16,52], or hybrid schemes [67]. Relevant wavelength selection in the context of spectral rendering has also been studied [76,42].…”

“…Our rendering process can be divided in three stages, the same way Sun decomposed his framework in [67]. First a spectral image is generated using spectral physical properties of the light sources and the surfaces of the scene (reflectances, BRDFs).…”

Techniques of Facial Synthesis: A Comprehensive Literature Review

“…This area has widespread applications in industry and scientific research including 3D design, computer animation, scientific visualization and virtual reality. In the last twenty-five years, a large number of research papers have been devoted to this area [3]. Nowadays, as people are enjoying the pleasure of realistically duplicating objects and complex scenes in the virtual world, synthesis of the human face has become one of the most promising research fields, because the face appears as the most perceptual knowledge of human beings [2].…”

“…These methods can be classified into three categories: anatomy based methods, geometry based methods and learning methods. Anatomy based methods build face models by estimating the dynamic facial muscle contractions from a sequence of human face images [3]. It needs a lot of pre-processing, such as registration of corresponding muscle points and setting constraints of muscle contraction to every sample.…”