Vector Faculty Affiliate wins top honours at CVPR 2019

July 9, 2019

Vector Faculty Affiliate Kyros Kutulakos has won the Best Paper Award at CVPR 2019. 

Kutulakos, along with his co-authors Shumian Xin, Sotiris Nousias, Aswin C. Sankaranarayanan, Srinivasa G. Narasimhan, and Ioannis Gkioulekas were awarded the conference’s top honor for their paper “A Theory of Fermat Paths for Non-Line-of-Sight Shape Reconstruction.” 

The paper proposes a novel theory to infer the shape of items concealed by corners, essentially allowing their algorithm to “see” around corners. It is a significant step to more accurately being able to mitigate non-line-of-sight reconstruction errors. 

Kutulakos is a professor of Computer Science at the University of Toronto and Vector Faculty Affiliate whose work is primarily focused on computer vision and computer graphics. 

CVPR is one of the world’s biggest computer vision conferences. This year’s event, which ran from June 16-20, was held in Long Beach, CA.


Standardized protocols are key to the responsible deployment of language models

The known unknowns: Vector researcher Geoff Pleiss digs deep into uncertainty to make ML models more accurate

A man looks at a white board with red formulas on it
Trustworthy AI

How to safely implement AI systems