I am a Research Scientist at Apple AI/ML. I am also an affiliate associate professor of Computer Science and Engineering at the University of Washington. My research combines deep learning, computer graphics, and computer vision to create usable systems that allow people to better capture, understand, edit, and visualize the world.
I collaborate with the University of Washington Graphics and Imaging Laboratory, Reality Lab, and Facial Expression Research Group where we combine artistic insight and deep learning techniques to better understand facial expressions and create amazing tools for artists.
Previously, I was a Research Scientist at Zillow where I helped create the state of the art in home interior capture and visualization. As a researcher at Amazon, I focused on 3D object creation from photos. My Ph.D. work focused on interactive image-based modeling systems for architectural structures. I spent 10 years at Microsoft Research working on a variety of projects ranging from creating a platform for building virtual worlds to implementing microphone arrays.
Generative Multiplane Images (GMPI) ECCV 2022 oral
"What is really needed to make an existing 2D GAN 3D-aware?"
To answer this question, we modify a classical GAN, i.e., StyleGANv2, as little as possible. We find that only two modifications are absolutely necessary:
A multiplane image style generator branch which produces a set of alpha maps conditioned on their depth;
A pose-conditioned discriminator.
Fast and Explicit Neural View Synthesis
Pengsheng Guo, Miguel Angel Bautista, Alex Colburn, Liang Yang, Daniel Ulbricht, Joshua M. Susskind, and Qi Shan
WACV 2022: https://arxiv.org/pdf/2107.05775.pdf
We study the problem of novel view synthesis from sparse source observations of a scene comprised of 3D objects. We propose a simple yet effective approach that is neither continuous nor implicit, challenging recent trends on view synthesis