Hello, I'm Keunhong Park

I am a research scientist at Google, where I work on computer vision, graphics, and machine learning.

Previously, I was a Ph. D student at the University of Washington advised by Ali Farhadi and Steve Seitz.

News

May 2022: I have defended and will be joining Google as a research scientist.
Oct 2021: FiG-NeRF has been accepted to 3DV 2021.
Sep 2021: HyperNeRF has been accepted to SIGGRAPH Asia 2021. We have also released the code.
Jul 2021: Nerfies has been accepted to ICCV 2021 as an oral presentation.
Jun 2021: We present HyperNeRF, which uses ideas from level set methods to model topologically changing scenes.
Nov 2020: We present Nerfies, a way to turn casually captured selfie videos into free-viewpoint visualizations.
Jun 2020: We’ve released the code and data for LatentFusion.
Feb 2020: LatentFusion, our paper on zero-shot pose estimation, has been accepted to CVPR 2020.

Keunhong Park, Utkarsh Sinha, Peter Hedman, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Ricardo Martin-Brualla, Steven M. Seitz

SIGGRAPH Asia, 2021

By applying ideas from level set methods, we can represent topologically changing scenes with NeRFs.

Christopher Xie, Keunhong Park, Ricardo Martin-Brualla, Matthew A. Brown

3DV, 2021

Given a lot of images of an object category, you can train a NeRF to render them from novel views and interpolate between different instances.

Keunhong Park, Utkarsh Sinha, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Steven M. Seitz, Ricardo Martin-Brualla

ICCV, 2021 Oral Presentation

Learning deformation fields with a NeRF let's you reconstruct non-rigid scenes with high fidelity.

Keunhong Park, Utkarsh Sinha, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Steven M. Seitz, Ricardo Martin-Brualla

CVPR, 2020

By learning to predict geometry from images, you can do zero-shot pose estimation with a single network.

Keunhong Park, Kostas Rematas, Ali Farhadi, Steven M. Seitz

SIGGRAPH Asia, 2018 Journal Cover

By pairing large collections of images, 3D models, and materials, you can create thousands of photorealistic 3D models fully automatically.