I have completed my Ph.D in 2016. I am now working as a research scientist at light. During my Ph.D, I have worked on example based image restoration and synthesis problems with my advisor James Hays. In 2014, I spent the summer with the Graphics group at Microsoft Research, Redmond. In 2012, I had the opportunity to work with Jue Wang and Sunghyun Cho as a summer intern at Adobe Research, Seattle.
My research interest lies in data-driven methods in computational photography and computer vision, with a focus on developing example-based, expressive image priors suitable for restoration and synthesis tasks.
Dissertation
The Devil is in the Details: Example-based Image Restoration and Detail Synthesis
[pdf, 147MB]
Libin Sun, Brown University, April ss2016
PhD Research
Super-resolution Using Constrained Deep Texture Synthesis
Libin Sun, James Hays
arXiv, 2017
[pdf]
*This work was done before early 2016 towards my dissertation (Chapter 6).
Lens Factory: Automatic Lens Generation Using Off-the-shelf Components
Libin Sun, Brian Guenter, Neel Joshi,
Patrick Therien, James Hays
arXiv, 2015
[pdf]
Good Image Priors for Non-blind Deconvoluton: Generic vs Specific
Libin Sun, Sunghyun Cho, Jue Wang, James Hays
European Conference on Computer Vision 2014 (ECCV)
[project page] |
[pdf] |
[poster]
Edge-based Blur Kernel Estimation Using Patch Priors
Libin Sun, Sunghyun Cho, Jue Wang, James Hays
IEEE International Conference on Computational Photography 2013 (ICCP)
[project page] |
[pdf] |
[slides]
Super-resolution from Internet-scale Scene Matching
Libin Sun, James Hays
IEEE International Conference on Computational Photography 2012 (ICCP)
[project page] |
[pdf] |
[slides] |
[talk video]
Previous Research
Large Scale kNN using CUDA and MPI
Hybrid MPI and GPU Approach to Efficiently Solving Large kNN Problem
Libin Sun, Cyrus Stoller and Tia Newhall, the 2010 TeraGrid Conference [poster]
[featured article] in International Science Grid This Week (ISGTW), Sep-15-2010