I am a computer science Ph.D. candidate at the Harvard School of Engineering and Applied Sciences, working with Todd Zickler. I am also currently a student researcher at Google Research, working with Jon Barron, Pratul Srinivasan, Peter Hedman, and Ben Mildenhall.
Research Focus:
I am interested in computer vision algorithms that are informed by geometry, physics, and graphics. By incorporating domain-specific constraints into differentiable architectures, I aim to improve the efficiency, generalizability , and interpretability of artificial vision systems.
Selected Projects:
 |
Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields
CVPR, 2022 (Oral Presentation, Best Student Paper Honorable Mention)
Neural radiance fields (NeRF) often produces incorrect specular highlights. We modify NeRF's representation of view-dependent appearance to fix that problem, and recover accurate surface normals. Our method also enables view-consistent scene editing.
|
 |
Field of Junctions: Extracting Boundary Structure at Low SNR
Dor Verbin, Todd Zickler
ICCV, 2021
By modeling each patch in an image as a generalized junction, our model uses concurrencies between different boundary elements such as junctions, corners, and edges, and manages to extract boundary structure from extremely noisy images where previous methods fail.
|
 |
Toward a Universal Model for Shape from Texture
We formulate the shape from texture problem as a 3-player game. This game simultaneously estimates the underlying flat texture and object shape, and it succeeds for a large variety of texture types.
|
 |
Unique Geometry and Texture from Corresponding Image Patches
We present a simple condition for the uniqueness of a solution to the shape from texture problem. We show that in the general case four views of a cyclostationary texture satisfy this condition and are therefore sufficient to uniquely determine shape.
|