Carnegie Mellon University

Group photo from the CVPR conference

June 20, 2019

Sankaranarayanan and co-authors win best paper at CVPR

ECE Associate Professor Aswin Sankaranarayanan and co-authors, Shumian Xin, Ioannis Gkioulekas, and Srinivasa Narasimhan from the Robotics Institute, along with colleagues from the University of Toronto and University College London, have received the Best Paper Award at the 2019 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) held in Long Beach, California. Their paper, titled, “A Theory of Fermat Paths for Non-Line-of-Sight Shape Reconstruction,” presents a novel theory of Fermat paths of light between a known visible scene and an unknown object not in the line of sight of a transient camera. These light paths either obey specular reflection or are reflected by the object’s boundary, and hence encode the shape of the hidden object.

The CVPR awards committee stated, "This paper makes significant advances in non-line-of-sight reconstruction -- in essence the ability to see around corners.  It is both a beautiful paper theoretically as well as inspiring.  It continues to push the boundaries of what is possible in computer vision." 

Read the full paper here.

Read more about CVPR here.