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. Most recently, I spent 1.5 years at NVIDIA AI working with Koki Nagano.

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

I'm currently at Google DeepMind in San Francisco as a Student Researcher with Pratul Srinivasan and Ben Poole.

Email / CV / Twitter / Scholar / GitHub

Research

I'm interested in machine learning and computer vision, especially making or leveraging generative priors for anything 3D.

	RealmDreamer: Text-Driven 3D Scene Generation with Inpainting and Depth Diffusion Jaidev Shriram, Alex Trevithick*, Lingjie Liu, Ravi Ramamoorthi International Conference on 3D Vision (3DV)*, 2025 project page / paper / code Sample 3D scenes from text with diffusion.
	What You See Is What You GAN: Rendering Every Pixel for High-Fidelity Geometry in 3D GANs Alex Trevithick, Matthew Chan, Towaki Takikawa, Umar Iqbal, Shalini De Mello, Manmohan Chandraker, Ravi Ramamoorthi, Koki Nagano Computer Vision and Pattern Recognition (CVPR), 2024 project page / paper Photorealistic geometry in 3D generative models by rendering every pixel.
	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 / two minute papers 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.