Eric Kee

I am currently at Adobe, on Marc Levoy's new team, where we are building computational photography methods for mobile devices. Previously I was at Uber ATG R&D led by Raquel Urtasun, where I developed methods for vehicle perception in driverless cars. Prior to that, I was a research scientist and engineer at Avametric, a physical simulation startup with Prof. James O'Brien, UC Berkeley. Prior to that, I was a research scientist and engineer at Facebook, computational imaging. Prior to that, I was a postdoc with professor Shree Nayar in the CAVE lab at Columbia. Lastly, and most importantly, my dissertation work was at Dartmouth with advisor Hany Farid.

My research background includes topics in: computer vision, machine learning, computer graphics, and image forensics.

If you would like to contact me: erickee AT gmail dot com

Here is my CV and google scholar link.

publications    (illustrated list)

Realistic Saliency Guided Image Enhancement.
  M. Miangoleh, Z. Bylinskii, E. Kee, E. Shechtman, Y. Aksoy
  CVPR 2023
 
Efficient Sensory Coding of Multidimensional Stimuli
  T. Yerxa, E. Kee, M. DeWeese, E. Cooper.
  PLOS Computational Biology, 2020 (see also ArXiV 2016)
 
Universal Embeddings for Spatio-Temporal Tagging of Self-Driving Logs
  S. Segal, E. Kee, W. Luo, A. Sadat, E. Yumer, R. Urtasun
  CoRL, 2020
 
DeepSignals: Predicting Intent of Drivers Through Visual Attributes
  D. Frossard, E. Kee, R. Urtasun
  ICRA, 2019
 
LaserNet: An Efficient Probabilistic 3D Object Detector for Autonomous Driving
  G. Meyer, A. Laddha, E. Kee, C. Vallespi-Gonzalez, C. Wellington
  CVPR, 2019
 
Uniform Transformation of Non-Separable Probability Distributions
  E. Kee
  ArXiV, 2016
 
Exposing Photo Manipulation From User-Guided 3-D Lighting Analysis
  T. Carvalho, H. Farid, E. Kee
  SPIE, 2015
 
Exposing Photo Manipulation from Shading and Shadows
  E. Kee, J. O'Brien, H. Farid
  Transactions on Graphics / SIGGRAPH, 2014
 
Photo Forensics from Partial Constraints
  E. Kee   (advisor: H. Farid)
  Ph.D. Dissertation, Dartmouth College, 2013
 
Exposing Photo Manipulation with Inconsistent Shadows
  E. Kee, J. O'Brien, H. Farid
  Transactions on Graphics / SIGGRAPH, 2013
 
A Perceptual Metric for Photo Retouching
  E. Kee, H. Farid
  Proceedings of the National Academy of Sciences, 2011
 
Digital Image Authentication from JPEG Headers
  E. Kee, M. K. Johnson, H. Farid
  IEEE Transactions on Information Forensics and Security, 2011
 
Modeling and Removing Spatially-Varying Optical Blur
  E. Kee, S. Paris, S. Chen, J. Wang
  IEEE International Conference on Computational Photography, 2011
 
Exposing Digital Forgeries from 3-D Lighting Environments
  E. Kee, H. Farid
  IEEE International Workshop on Information Forensics and Security, 2010
 
Digital Image Authentication from Thumbnails
  E. Kee, H. Farid
  SPIE Symposium on Electronic Imaging, San Jose, CA, 2010
 
Detecting Photographic Composites of Famous People
  E. Kee, H. Farid
  Technical Report, TR2009-656, Dartmouth College, Computer Science
 
A novel approach for semi-automated segmentation of MS lesions on FLAIR imaging: Reliability and clinical correlates
  H. Wishart, E. Kee, J. Ford, J. MacDonald, S. Aney, S. LoVerde
  Annual Meeting of the International Neuropsychological Society, Acapulco, Mexico, February 3-6, 2010
 
Printer Profiling for Forensics and Ballistics
  E. Kee, H. Farid
  ACM Multimedia and Security Workshop, Oxford, UK, 2008
 
An Adaptive Genetic Algorithm
  E. Kee, S. Airey
   Proceedings of the Genetic and Evolutionary Computation Conference, 2001

press

for "Exposing Photo Manipulation with Inconsistent Shadows"
New York Times
MIT Technology Review
Live Science, via Mashtable
Scientific American

for "A Perceptual Metric for Photo Retouching"
New York Times
Wired
The Economist
Nature
New Scientist
Marketplace
Kojo
The Guardian
CBS
MSNBC