Hyo Jin Kim

hyojin (at) cs.unc.edu

Facebook Reality Labs

I am a Research Scientist at Facebook Reality Labs Research. I received my PhD degree in Computer Science at UNC Chapel Hill, advised by Prof. Jan-Michael Frahm in August 2018.

My research focuses on computer vision and machine learning. I am especially interested in visual search, including large-scale image retrieval and image geo-localization (a.k.a. place recognition), as well as recognition problems such as scene recognition. I like to apply data-driven approaches, and leverage weak labels for minimizing human supervision.

I received M.S. degree in EECS at Seoul National University under the supervision of Prof. Kyoung Mu Lee in 2012, and B.S. in Electrical Engineering at KAIST in 2010. I was an intern at NEC Labs. Before coming to Chapel Hill, I also worked as a researcher at KIST and Pusan National University in Korea, working on object recognition and speech recognition, respectively.

Publications

Google Scholar

International Conference Paper

[1] NinjaDesc: Content-Concealing Visual Descriptors via Adversarial Learning
Tony Ng, Hyo Jin Kim (corresp. author), Vincent Lee, Daniel Detone, Tsun-Yi Yang, Tianwei Shen, Eddy Ilg, Vassileios Balntas, Krystian Mikolajczyk, Chris Sweeney
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans. June 2022.
[Paper](25.3% acceptance rate (2064/8161))

[2] Verifiable Access Control for Augmented Reality Localization and Mapping
Shaowei Zhu, Hyo Jin Kim , Maurizio Monge, G Edward Suh, Armin Alaghi, Brandon Reagen, Vincent Lee
(Under Submission) 2022
[Paper]

[3] Analysis and Mitigations of Reverse Engineering Attacks on Local Feature Descriptors
Deeksha Dangwal, Vincent T Lee, Hyo Jin Kim , Tianwei Shen, Meghan Cowan, Rajvi Shah, Caroline Trippel, Brandon Reagen, Timothy Sherwood, Vasileios Balntas, Armin Alaghi, Eddy Ilg
British Machine Vision Conference (BMVC), Online. September 2021.
[Paper]

[4] Domain Adaptation of Learned Features for Visual Localization
Sungyong Baik, Hyo Jin Kim, Tianwei Shen, Eddy Ilg, Kyoung Mu Lee, and Chris Sweeney
British Machine Vision Conference (BMVC), Online. September 2020.
[Paper]

[5] Hierarchy of Alternating Specialists for Scene Recognition
Hyo Jin Kim and Jan-Michael Frahm
European Conference on Computer Vision (ECCV), Munich, Germany. September 2018.
[Paper] [Supplementary]

[6] Learned Contextual Feature Reweighting for Image Geo-Localization
Hyo Jin Kim, Enrique Dunn, and Jan-Michael Frahm
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA. July 2017.
[Paper] [Supplementary] [Project Page] [Presentation] [Poster] [Code]
(Spotlight presentation, 8% acceptance rate (215/2680))

[7] Predicting Good Features for Image Geo-Localization Using Per-Bundle VLAD
Hyo Jin Kim, Enrique Dunn, and Jan-Michael Frahm
15th IEEE International Conference on Computer Vision (ICCV), Santiago, Chile. December 2015.
[Paper] [Supplementary] [Project Page] [Video Spotlight] [Poster]
(30.9% acceptance rate (525/1698))

International Workshops

[8] Symmetry-Growing for Skewed Rotational Symmetry Detection
Hyo Jin Kim, Minsu Cho, and Kyoung Mu Lee
Winner (Rotation Symmetry) of Symmetry Detection in Real-World Images - A Competition, in conjunction with 24th IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011.
[Preprint] [Slides] [Program] [Video]
(Oral presentation, Received travel funds from US NSF)

Thesis

[9] Learning Adaptive Representations for Image Retrieval and Recognition
Hyo Jin Kim
Doctoral Thesis,
Deptartment of Computer Science, University of North Carolina at Chapel Hill, USA. August 2018.
[PDF]

[10] A Skewed Rotational Symmetry Detection Technique using Symmetry-Growing
Hyo Jin Kim
Masters Thesis,
Deptartment of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea. January 2012.
[PDF]

More about me.

Media Coverage

Computer Vision News (Nov. 2019) Women in Computer Vision Series [Link]

Honors and Awards

Winner of Rotation Symmetry Detection, IEEE International Workshop on Symmetry Detection in Real World Images - A Competition in conjunction with CVPR (2011) (Received travel funds from US NSF)
Bronze Prize, 18-th Samsung Humantech Thesis Award (2012) (Acceptance rate = 5.1% with a total of 1462 submitted papers)
Travel Award from Women in Computer Vision Workshop (WiCV) in conjunction with CVPR (2017)
Scholarship for Superior Academic Performance (2010)
Merit-based National Science and Engineering Undergraduate Scholarship (2006-2009)
Bronze Prize, National Chemistry Frontier (2005)
Honor Prize, Korean Chemistry Olympiad (2004)

Teaching

TA, Scientific Computing (COMP 116), UNC Chapel Hill, 2016 Fall
TA, Linear Algebra for Electrical Systems (EE210), Seoul National University, 2010 Fall
Grader, Communication Engineering (EH26218), Pusan National University, 2012 Spring

Advanced Courses

Machine Learning, Probabilistic Reasoning, Random Signal Theory, Detection and Estimation, Optimization, Image Processing and Analysis, Digital Speech Processing, Computer Vision, Display Engineering, Visual Solid Shapes, Computational Photography

Technical

Language: Python, Matlab, C/C++, Lua, Java
Library: Pytorch, Torch, Caffe, OpenCV

Contact Information

Email : hyojin (at) cs.unc.edu

You can also find me on...