Date Posted: 2/01/2022

Zekun Hao, a doctoral candidate studying under the direction of Professor Serge Belongie, has received an NVIDIA Graduate Fellowship for 2022-23, which will support his continued pursuit of GPU (Graphics Processing Unit)-related research.

First, congratulations on your 2022-23 NVIDIA Graduate Fellowship Award. Please tell us about your current research and what you hope to do with your award.

My research focuses on controllable neural synthesis of photorealistic 3D worlds. The fellowship will allow me to better dedicate myself to this line of research. The award also comes with a GPU, which will allow me to try out new ideas faster.

Can you briefly unpack the potential significance of your research? What do you hope to accomplish with this line of investigation?

Creating photorealistic 3D worlds is a painstaking task that requires substantial skill, time, and resources. My research has the potential of greatly reducing the effort and skill needed for such tasks, allowing professional and even casual creators to create their imaginary 3D worlds effortlessly. This will be accomplished by an algorithm learning the essence of photorealism from real-world images and then applying the knowledge to a simple and abstract user input, converting it to a highly detailed photorealistic 3D world.

As a preliminary attempt on this problem, GANcraft, my previous work in collaboration with NVIDIA, has already achieved encouraging results. GANcraft converts a user-provided sparse voxel world, such as the one used in the Minecraft game, to photorealistic 3D worlds with a variety of appearances.

What is a promising outcome of your research that you hope to see?

Just like how the discovery of perspective projection has transformed Renaissance paintings, and how the availability of computers has given rise to digital arts, I hope my research will become an invaluable tool that enables new styles of artworks.

What’s your experience doing research at Cornell been like?

As a Ph.D. student at Cornell, I really enjoyed the open and collaborative atmosphere—in fact, many of my ideas are sparked by the discussions and collaborations with my colleagues as well as industry collaborators. Cornell is also a very supportive place, making me feel confident and comfortable to try out challenging and risky projects. (And BTW, I really enjoyed the convenience of living in NYC that Cornell Tech has offered :) )

For more than two decades, NVIDIA has supported graduate students doing GPU-based work through the NVIDIA Graduate Fellowship Program.

Selected from a highly competitive applicant pool, the awardees will participate in a summer internship preceding the fellowship year. The work they’re doing puts them at the forefront of GPU computing, with fellows tackling projects in deep learning, robotics, computer vision, computer graphics, architecture, circuits, high performance computing, life sciences and programming systems.

“Our fellowship recipients are among the most talented graduate students in the world,” said NVIDIA Chief Scientist Bill Dally. “They’re working on some of the most important problems in computer science, and we’re delighted to support their research.”

The NVIDIA Graduate Fellowship Program is open to applicants worldwide.