Gene Chou

I am a research assistant in the Princeton Computational Imaging Lab, advised by Professor Felix Heide. I received my bachelor's in computer science and math from Princeton University and I am applying to PhD programs this year. My research interests are in neural rendering and 3D scene representations.

Contact: gchou@princeton.edu

CV  /  Github  /  LinkedIn


Research

I'm currently working on the generalization of object representations, specifically through neural signed distance functions.


DiffusionSDF: Conditional Generative Modeling of Signed Distance Functions
Gene Chou, Yuval Bahat, Felix Heide
Arxiv preprint
[Paper] [Project Page] [Teaser Video] [Code]

This work proposes a probabilistic generative model that generates clean and diverse 3D meshes conditioned on partial point clouds, single 2D images, and real-scanned point clouds. We also solve a learning problem of diffusing the weights of implicit neural functions while providing geometric guidance.

GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions
Gene Chou, Ilya Chugunov, Felix Heide
NeurIPS 2022 (Featured)
[Paper] [Code] [Project page]

Neural signed distance functions (SDFs) are a compact and versatile way of representing 3D objects, but state-of-the-art methods for SDF estimation struggle to fit more than a few shapes at a time. This work presents a two stage semi-supervised meta-learning approach that learns generic shape priors to reconstruct over a hundred unseen object classes.

Curriculum Vitae

Website template borrowed from Jon Barron.