|
Zach Salehe I'm a second-year computer science PhD student at Harvard advised by Todd Zickler. Before Harvard, I did my undergrad in computer science at the University of Toronto, where my research was advised by Kyros Kutulakos and David Lindell. |
|
ResearchI'm broadly interested in computational imaging, computer vision, and computer graphics. My current research focuses on inverse rendering for non-line-of-sight reconstruction. |
|
Dark3R: Learning SfM in the Dark
Andrew Y. Guo, Anagh Malik, SaiKiran Tedla, Yutong Dai, Yiqian Qin, Zach Salehe, Benjamin Attal, Sotiris Nousias, Kiriakos N. Kutulakos, David B. Lindell CVPR, 2026 (Highlight) project page / arXiv / code We provide a framework for structure from motion in dark (< -4dB SNR), and use the resulting camera poses to enable volumetric denoising via radiance field reconstruction. |
|
A Foundation Model for Soccer
Ethan Baron*, Daniel Hocevar*, Zach Salehe* arXiv, 2024 arXiv / code We use a transformer decoder architecture to develop a foundation model for soccer actions, capable of predicting the next action following from a given input sequence. |
 
|
Develping a Simulation Pipeline for Next-Generation Computational Cameras
Zach Salehe, Rahul Gulve, Esther Y. H. Lin, David B. Lindell, Kiriakos N. Kutulakos 2023 code We provide a simulation pipeline in order to accurately model the T6, a next-generation computational camera developed by the University of Toronto. |
|
Template stolen from Jon Barron. |