Eric Ming Chen
I’m Eric, a junior at Cornell University studying Computer Science! My interests include 3D computer vision, geometric deep learning and neural fields. I conduct research with Hadar Averbuch-Elor, Jin Sun and Noah Snavely at Cornell Tech; Abe Davis at Cornell; and with CUAI.
Email: emc348[at]cornell.edu Github: echen01
Research
Computer Vision
I’m interested in many areas of computer vision, many of which deal with content creation.
Light Fields
- Light fields are a classic scene representation for image based rendering. While light fields have seemingly fallen by the wayside as radiance fields have gained popularity, I think the fact that they allow for real-time view synthesis is still useful.
- I am actively investigating how light fields can be used for content generation.
Transforming Faces Through Time
- Have you ever imagined how Charlie Chaplin would have looked like today, or how a celebrity would have looked in the 1920s? We take a data driven approach to picture how a person would look in another decade’s style.
- We introduced a new image dataset consisting of over 25k historical people taken from the Wikimedia Commons.
Geometric Machine Learning
I am also interested in discovering where geometry and symmetry play unexpected roles in data.
Riemannian Residual Neural Networks
- We introduce a simple and general way to design residual neural networks on smooth manifolds.
- We apply our model to show improved performance on hyperbolic graph embedding, and covariance matrix classification.
Publications and Preprints
What’s in a Decade? Transforming Faces Through Time
Eric M. Chen, Jin Sun, Apoorv Khandelwal, Dani Lischinski, Noah Snavely, Hadar Averbuch-Elor
Eurographics 2023
Website Paper
Riemannian Residual Neural Networks
Isay Katsman*, Eric M. Chen*, Sidhanth Holalkere*, Anna Asch*, Aaron Lou, Ser-Nam Lim, Christopher De Sa
ICML 2022 Workshop on Topology, Algebra, and Geometry in ML
Paper Poster
* Equal Contribution
Writings
The elegant geometry behind Rwanda’s COVID-19 pooled testing strategy
Eric M. Chen, Medha Bulumulla, Megan Rochlin
Cornell Data Journal, Published Dec. 18, 2020
Article
Projects
Cornell Rendering Competition Submission
I was the runner up to the Cornell CS 6630 rendering competition. I won a copy of Fundamentals of Computer Graphics signed by Steve Marschner!
Slides