Project Grant 2212085

This four-year, $400,000 Project Grant from the National Science Foundation's Computer and Information Science and Engineering program will fund the development of new neural material models for more realistic computer graphics at the University of California, San Diego. Specifically, the university researchers will build machine learning-based models to more accurately simulate the complex reflectance patterns of materials at fine scales, going beyond current parametric models. The neural materials are intended to support a wide range of computer graphics applications from visual effects to virtual reality by accommodating surface fuzziness, translucency, and fine-scale geometric details not captured by existing models. If successful, these new material representations could transform how textures and surfaces are modeled across various industries utilizing computer-generated imagery.