James Fong

I am a computer science PhD student at UC Berkeley, advised by Ren Ng. I have broad research interests including computer graphics, computational photography, and human vision. My most recent work is focused on understanding the earliest stages of the human visual system through cell-by-cell psychophysics. In my free time, I enjoy video game development and 3D animation.

profile photo
Publications
Novel Color via Stimulation of Individual Photoreceptors at Population Scale
James Fong*, Hannah K. Doyle*, Congli Wang*, Alexandra E. Boehm, Sofie R. Herbeck, Vimal Prabhu Pandiyan, Brian P. Schmidt, Pavan Tiruveedhula, John E. Vanston, William S. Tuten, Ramkumar Sabesan, Austin Roorda, Ren Ng
Science Advances, 2025

We introduce a principle, Oz, for displaying color imagery: directly controlling the human eye's photoreceptor activity via cell-by-cell light delivery. On this display, concentrating light into the M-type cone cells produces a novel color outside the natural human gamut. Subjects describe the color as a blue-green of unprecedented saturation, and indicate that it is more saturated than a near-monochromatic laser of equivalent hue.

Characterizing Cone Spectral Classification by Optoretinography
Vimal Prabhu Pandiyan, Sierra Schleufer, Emily Slezak, James Fong, Rishi Upadhyay, Austin Roorda, Ren Ng, Ramkumar Sabesan
Biomedical Optics Express, 2022

Thorough evaluation of optoretinography (ORG) in comparison to densitometry, which are two methods for determining the L/M/S spectral type of cone cells in awake humans. My primary role in this project was the development of the Retina Map Alignment Tool for efficiently finding corresponding cones between the datasets generated by each method.

R-SLAM: Optimizing Eye Tracking from Rolling Shutter Video of the Retina
Jay Shenoy*, James Fong*, Jeffrey Tan, Austin Roorda, Ren Ng
ICCV, 2021
project page / video

Computational imaging-based method for jointly estimating the motion and appearance of the human retina using data collected on an adaptive optics scanning laser ophthalmoscope.

How to See Impossible Colors: First Steps Toward the Oz Vision Display
James Fong
Master's Thesis, 2021

In collaboration with the folks at Roorda Lab, I was able to display world-fixed squares of cyan and orange with only green laser light. I developed a high-performance C++/CUDA software layer to enable per-photoreceptor computer graphics via an adaptive optics scanning laser ophthalmoscope.

Experience

Graduate Student Researcher | 2021–Present
Ng Lab and Roorda Lab, UC Berkeley

Computer Vision, ML & Computational Photography Research Intern | Summer 2022
Adobe Inc.

Undergraduate Student Instructor | Spring 2019
Computer Graphics and Imaging Course, UC Berkeley

  • Designed a new GLSL shaders section of a class project.
  • Leaded port of 3 previously Linux-only C++ projects to Windows.
  • Advised on the final class project to students.

Software Developer (volunteer) | 2017–2018
Pioneers in Engineering, volunteer student outreach organization

  • Designed simple coding challenges for competition participants.
  • Wrote automated tests for participant's implementations.
  • Assisted in student recruiting.

Student Intern, IT Programmer | Summer 2014, Summer 2015
Bureau of Street Use and Mapping, City of San Francisco

  • Assisted IT Manager with database applications.
  • Prototype for street inspector tablet app and permit database map visualizer.
Class Projects
Perspective-Aware Collaging Tool
CS 294-26: Computational Photography, Spring 2020
project page / video / code

Tool for easily making photo collages with accurate vanishing point geometry.

Raymarched Microgeometry on Triangle Meshes
CS 184: Computer Graphics, Spring 2018
project page / video

Real-time method for adding extra detail onto mesh surfaces. This is similar to traditional displacement mapping, but can handle overhangs and does not rely on tessellation.
(Final Class Project Showcase Winner)


Site template courtesy of Jon Barron