Paul Upchurch

PhD Candidate
Computer Science Department
Cornell University

345 Gates Hall
paulu at

About me

I am a PhD candidate at Cornell University working in the Graphics and Vision Group. My advisors are Kavita Bala, Noah Snavely and Kilian Q. Weinberger. My research interests are in computer vision, machine learning, computer graphics, and human computation. My work is on data-driven models for understanding and editing appearance in photographs and new approaches for labeling images with crowdsourced workers.

[ Google Scholar ]


Deep Feature Interpolation For Image Content Changes
The IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017)
*Authors contributed equally
Interactive Consensus Agreement Games For Labeling Images
The AAAI Conference on Human Computation and Crowdsourcing (HCOMP 2016)
Material Recognition in the Wild with the Materials in Context Database
The IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2015)
*Authors contributed equally
Reasoning about Photo Collections using Models of Outdoor Illumination
British Machine Vision Conference (BMVC 2014)
OpenSurfaces: A Richly Annotated Catalog of Surface Appearance
ACM Transactions on Graphics (SIGGRAPH 2013)
Tightening the Precision of Perspective Rendering
Journal of Graphics Tools (JGT 2012)

Technical Reports

Deep Manifold Traversal: Changing Labels with Convolutional Features
arXiv 2016
*Authors contributed equally
From A to Z: Supervised Transfer of Style and Content Using Deep Neural Network Generators
arXiv 2016