Alex Trevithick

I am a CSE PhD student at UC San Diego, advised by Professor Ravi Ramamoorthi. In the past, I have had the fortune to work at the Max Planck Institute and the University of Oxford. Now, I am interning at NVIDIA AI through Summer 2023.

I'm lucky to be supported by an NSF Graduate Research Fellowship, Jacobs School of Engineering Fellowship, and SHORE Fellowship.

Email  /  CV  /  Twitter  /  Scholar  /  GitHub

profile photo
Research

I'm interested in machine learning and computer vision, especially visual intelligence and representing the 3D world from one or few images.

AI-mediated 3D Videoconferencing
Michael Stengel, Koki Nagano Chao Liu, Matthew Chan, Alex Trevithick, Shalini De Mello, Jonghyun Kim, David Luebke, Amrita Mazumdar, Shengze Wang, Mayoore Jaiswal
SIGGRAPH Emerging Technologies, 2023
project page / paper

Real-time demo for immersive 3D videoconferencing built with Live 3D Portrait.

Real-Time Radiance Fields for Single-Image Portrait View Synthesis
Alex Trevithick, Matthew Chan, Michael Stengel, Eric R. Chan, Chao Liu, Zhiding Yu, Sameh Khamis, Manmohan Chandraker, Ravi Ramamoorthi, Koki Nagano
ACM Transactions on Graphics (SIGGRAPH), 2023
project page / paper

Real-time encoding and view synthesis from a single portrait image learned with strictly synthetic data.

NerfDiff: Single-image View Synthesis with NeRF-guided Distillation from 3D-aware Diffusion
Jiatao Gu, Alex Trevithick, Kai-En Lin, Josh Susskind, Christian Theobalt, Lingjie Liu, Ravi Ramamoorthi
International Conference on Machine Learning (ICML), 2023
project page / paper / code

Distill the knowledge of a 3D-aware conditional diffusion model into a triplane NeRF.

PVP: Personalized Video Prior for Editable Dynamic Portraits using StyleGAN
Kai-En Lin, Alex Trevithick, Keli Chang, Michel Sarkis, Mohsen Ghafoorian, Ning Bi, Gerhard Reitmayr, Ravi Ramamoorthi
Computer Graphics Forum (EGSR), 2023
project page / paper

Leverage the StyleGAN latent space for multi view consistent edits in real-time.

GRF: Learning a General Radiance Field for 3D Scene Representation and Rendering
Alex Trevithick, Bo Yang
International Conference on Computer Vision (ICCV), 2021
paper / code / video

Per-pixel features improve NeRF and allow it to generalize to new scenes without retraining.

Updates

6/19/23: Starting second internship at NVIDIA AI.

6/8/23: Giving talk at the annual UCSD visual computing retreat.

5/29/23: Jensen presents our work at Computex!

5/26/23: Giving talk at Google Labs on Live 3D Portrait.

Teaching
Teaching Assistant in Measure Theory and Hilbert Spaces, Fall 2020

Teaching Assistant in Computational Linear Algebra, Fall 2018
Thanks to Jon Barron for this awesome template!