Real-time realistic skin translucency
Diffusion theory allows the production of realistic skin renderings. The dipole and multipole models allow for solving challenging diffusion-theory equations efficiently. By using texture-space diffusion, a Gaussian-based approximation, and programmable graphics hardware, developers can create real-...
| Publié dans: | IEEE computer graphics and applications. - 1991. - 30(2010), 4 vom: 15. Juli, Seite 32-41 |
|---|---|
| Auteur principal: | |
| Autres auteurs: | , , |
| Format: | Article en ligne |
| Langue: | English |
| Publié: |
2010
|
| Accès à la collection: | IEEE computer graphics and applications |
| Sujets: | Journal Article Research Support, Non-U.S. Gov't |
| Résumé: | Diffusion theory allows the production of realistic skin renderings. The dipole and multipole models allow for solving challenging diffusion-theory equations efficiently. By using texture-space diffusion, a Gaussian-based approximation, and programmable graphics hardware, developers can create real-time, photorealistic skin renderings. Performing this diffusion in screen space offers advantages that make diffusion approximation practical in scenarios such as games, where having the best possible performance is crucial. However, unlike the texture-space counterpart, the screen-space approach can't simulate transmittance of light through thin geometry; it yields unrealistic results in those cases. A new transmittance algorithm turns the screen-space approach into an efficient global solution, capable of simulating both reflectance and transmittance of light through a multilayered skin model. The transmittance calculations are derived from physical equations, which are implemented through simple texture access. The method performs in real time, requiring no additional memory usage and only minimal additional processing power and memory bandwidth. Despite its simplicity, this practical model manages to reproduce the look of images rendered with other techniques (both offline and real time) such as photon mapping or diffusion approximation |
|---|---|
| Description: | Date Completed 28.10.2010 Date Revised 23.07.2010 published: Print Citation Status MEDLINE |
| ISSN: | 1558-1756 |
| DOI: | 10.1109/MCG.2010.39 |