Faculty Candidate -- Ling-Qi Yan

Date: 
Wednesday, April 4, 2018 - 3:30pm
Location: 
HFH 1132
Title: 
Physically-based Modeling and Rendering of Complex Visual Appearance
Host: 
Matthew Turk

Abstract: Photorealistic Rendering in Computer Graphics is increasingly important.

Whether in movies or video games, breathtaking graphics have become one of the most

crucial factors to their success. However, state of the art rendering still struggles with

two fundamental challenges -- realism and speed. It looks artificial and overly perfect

and is slow for both offline and interactive applications.

In this talk, I will first introduce my research on physically-based modeling and rendering

of visual appearance at real world complexity, which not only brings next generation

details to rendered images, but also greatly improves their realism. Specifically, I will

talk about 1) rendering glints from complex surfaces, such as metallic flakes and

scratches, to match the appearance of the imperfect real world and 2) modeling animal

fur from inside every fur fiber to achieve the overall furry and realistic appearance of

animals. After that, I will briefly introduce a sample-and-reconstruct approach to make

accurate rendering of complex effects real-time, by exploring Fourier analysis on light

transport and designing practical filtering algorithms on severely undersampled noisy

renderings.

Bio: Ling-Qi Yan is a fifth year Ph.D. candidate in the Department of Electrical

Engineering and Computer Sciences at UC Berkeley. His research is in Computer

Graphics, mainly aimed at rendering photo-realistic visual appearance at real world

complexity, building theoretical foundations mathematically and physically to reveal the

principles of the visual world. Ling-Qi's research interests include appearance modeling,

real-time ray tracing, sampling and reconstruction theory, volumetric scattering and light

transport algorithms.

Everyone welcome!