Gunhee Kim, PhD (김건희)

Gunhee Kim at DC 

Assistant Professor
Vision and Learning Lab
Department of Computer Science and Engineering
Seoul National University (SNU)

Rm 327 Bldg 302
1 Gwanak-ro, Gwanak-gu Seoul 08826 South Korea
Phone: +82 2 880 7300 (7289) Fax: +82 2 871 4912
Email: gunhee at { or}

Administrative Assistant: Jaeyoung Park (jaeyoung at

[Curriculum Vitae]

What's new?

To Prospective students

  • If you want to pursuit graduate study in our lab, you'd better first apply for internship. I recommend you to contact me at least one semester before your graduate application. Earlier would be better. Openings often close very early.

  • Usual openings per year: 2 Integrated PhD, 1 MS-only, and 0–1 PhD students.

  • For foreign students: Unless you already have valid Korean visa, it is very difficult for us to hire you as intern. Instead, I can accept 0–1 Korean government-funded foreign students [See details of KGSP].


Selected Projects

Gunhee Kim, ICML2017 SplitNet: Learning to Semantically Split Deep Networks for Parameter Reduction and Model Parallelization
  Juyong Kim, Yookoon Park, Gunhee Kim, and Sung Ju Hwang (UNIST)
  ICML 2017 [code] [pdf]
Attend-to-you, CVPR2017 Attend to You: Personalized Image Captioning with Context Sequence Memory Networks
  Cesc Chunseong Park, Byeongchang Kim, and Gunhee Kim
  CVPR 2017 (Spotlight) [code/dataset] [pdf]
TGFI-QA, CVPR2017 TGIF-QA: Toward Spatio-Temporal Reasoning in Visual Question Answering
  Yunseok Jang, Yale Song (Yahoo Research), YoungJae Yu, Youngjin Kim, and Gunhee Kim
  CVPR 2017 (Spotlight) [code/dataset] [pdf]
GazeCap, CVPR2017 Supervising Neural Attention Models for Video Captioning by Human Gaze Data
  YoungJae Yu, Jongwook Choi, Yeonhwa Kim, Kyung Yoo, Sang-Hun Lee, and Gunhee Kim
  CVPR 2017
LSMDC, CVPR2017 End-to-end Concept Word Detection for Video Captioning, Retrieval, and Question Answering
  YoungJae Yu, Hyungjin Ko, Jongwook Choi and Gunhee Kim
  CVPR 2017 (Spotlight)
  Winners of two tracks of LSMDC 2016 (Annotation and Retrieval, and Fill-in-the-Blank) [old paper] [presentation]
AAAI2017 Detection and Recognition of Text Embedded in Online Images via Neural Context Models
  Chulmoo Kang, Gunhee Kim, and Suk I. Yoo
  AAAI 2017 [code] [pdf]
ECCV2016 Taxonomy-Regularized Semantic Deep Convolutional Neural Networks
  Wonjoon Goo, Juyong Kim, Gunhee Kim, and Sung Ju Hwang (UNIST)
  ECCV 2016 [code] [pdf]
Poseidon, arXiv2015 Poseidon: A System Architecture for Efficient GPU-based Deep Learning on Multiple Machines
  Hao Zhang, Zhiting Hu, Jinliang Wei, Pengtao Xie (CMU), Gunhee Kim, Qirong Ho (I2R), and Eric P. Xing (CMU)
  ATC 2016 (Poster), ICML 2016 MLsys workshop, arXiv:1512.06216 [code] [wiki] [pdf]
Gunhee Kim, VR2015 A Calibration Method for Optical See-through Head Mounted Displays with a Depth Camera
  Hanseul Jun (SNU) and Gunhee Kim
  IEEE VR 2016(Acceptance = 25.9%) [project] [pdf]
Gunhee Kim, NIPS2015 Expressing an Image Stream with a Sequence of Natural Sentences
  Cesc Chunseong Park (SNU) and Gunhee Kim
  NIPS 2015(Acceptance = 403/1838~21.9%) [code] [pdf]
  A longer version in IEEE TPAMI [pdf]
Gunhee Kim, ICCV2015 Storyline Representation of Egocentric Videos and Its Applications to Story-based Search
  Bo Xiong (UT Austin), Gunhee Kim, and Leonid Sigal (Disney Research)
  ICCV 2015(Acceptance = 525/1698~30.9%)
Gunhee Kim, KDD2015 Dynamic Topic Modeling for Monitoring Market Competition from Online Text and Image Data
  Hao Zhang (CMU), Gunhee Kim, and Eric P. Xing (CMU)
  KDD 2015 (Research) (Acceptance = 159/819~19.4%)
Gunhee Kim, KDD2015 Discovering Collective Narratives of Theme Parks from Large Collections of Visitors’ Photo Streams
  Gunhee Kim and Leonid Sigal (Disney Research)
  KDD 2015 (Practice)
Gunhee Kim, CVPR2015 Ranking and Retrieval of Image Sequences from Multiple Paragraph Queries
  Gunhee Kim, Seungwhan Moon (CMU), and Leonid Sigal (Disney Research)
  CVPR 2015 (Acceptance = 602/2123~28.4%)
Gunhee Kim, CVPR2015 Joint Photo Stream and Blog Post Summarization and Exploration
  Gunhee Kim, Seungwhan Moon (CMU), and Leonid Sigal (Disney Research)
  CVPR 2015 (Acceptance = 602/2123~28.4%)
Gunhee Kim, CVPR2014 Reconstructing Storyline Graphs for Image Recommendation from Web Community Photos
  Gunhee Kim and Eric P. Xing (CMU)
  CVPR 2014 (Oral) (Acceptance = 104/1807~5.75%)
Gunhee Kim, CVPR2014 Joint Summarization of Large-scale Collections of Web Images and Videos for Storyline Reconstruction
  Gunhee Kim, Leonid Sigal (Disney Research), and Eric P. Xing (CMU)
  CVPR 2014 (Acceptance = 540/1807~29.88%)
Gunhee Kim, WSDM2014 Visualizing Brand Associations from Web Community Photos
  Gunhee Kim and Eric P. Xing (CMU)
  WSDM 2014 (Acceptance = 64/356~18.0%)
  A shorter version in LSVisCom (with ICCV 2013) (Best Paper Award)
  Press Release: [CMU News][Science Daily] [Register] [Popcity]
Gunhee Kim, CVPR2013 Jointly Aligning and Segmenting Multiple Web Photo Streams for the Inference of Collective Photo Storylines
  Gunhee Kim and Eric P. Xing (CMU)
  CVPR 2013 (Oral) (Acceptance = 60/1870~3.2%)
Gunhee Kim, WSDM2013 Time-Sensitive Web Image Ranking and Retrieval via Dynamic Multi-Task Regression
  Gunhee Kim and Eric P. Xing (CMU)
  WSDM 2013 (Acceptance = 73/387~18.9%)
Gunhee Kim, KDD2012 Web Image Prediction Using Multivariate Point Processes
  Gunhee Kim, Li Fei-Fei (Stanford), and Eric P. Xing (CMU)
  KDD 2012 (Research) (Acceptance = 133/755~17.6%)
Gunhee Kim, CVPR2012 On Multiple Foreground Cosegmentation
  Gunhee Kim and Eric P. Xing (CMU)
  CVPR 2012 (Oral) (Acceptance = 48/1933~2.48%)
Gunhee Kim, ICCV2011 Distributed Cosegmentation via Submodular Optimization on Anisotropic Diffusion
  Gunhee Kim, Eric P. Xing (CMU), Li Fei-Fei (Stanford), and Takeo Kanade (CMU)
  ICCV 2011 (Oral) (Acceptance = 45/1285~3.50%)
Gunhee Kim, ECCV2010 Modeling and Analysis of Dynamic Behaviors of Web Image Collections
  Gunhee Kim, Eric P. Xing (CMU) and Antonio Torralba (MIT)
  ECCV 2010 (Acceptance = 322/1174~27.4%)
Gunhee Kim, NIPS2009 Unsupervised Detection of Regions of Interest Using Iterative Link Analysis
  Gunhee Kim and Antonio Torralba (MIT)
  NIPS 2009 (Acceptance = 263/1105~23.8%)
Gunhee Kim, ACMMIR2009 Unsupervised Modeling and Recognition of Object Categories with Combination of Visual Contents and Geometric Similarity Links
  Gunhee Kim, Christos Faloutsos (CMU) and Martial Hebert (CMU)
  ACM MIR 2008 (Oral) (Acceptance =20/264~7.6%)
Gunhee Kim, CVPR2008 Unsupervised Modeling of Object Categories Using Link Analysis Techniques
  Gunhee Kim, Christos Faloutsos (CMU) and Martial Hebert (CMU)
  CVPR 2008 (Oral) (Acceptance =62/1593~3.9%)
Gunhee Kim, WACV2008 Detection of Salient Regions using Imagery and 3-D Data
  Gunhee Kim, Daniel Huber (CMU) and Martial Hebert (CMU)
  WACV 2008 (Oral)
Gunhee Kim, ICAR2008 Line Feature-based Object Recognition for Textureless Indoor Objects
  Gunhee Kim and Martial Hebert (CMU)
  ICAR 2008
Gunhee Kim, GSPN Navigation Behavior Selection using Generalized Stochastic Petri Nets (GSPN) (Sep 2003 ~ Feb 2005)
  with Woojin Chung (Korea Univ.)
Gunhee Kim, PSR Development of Multifunctional Service Robots (Aug 2001 ~ Aug 2003)
  with Woojin Chung (Korea Univ.)



Instructor at Seoul National University for:

  • 4190.773 Probabilistic Graphical Models, 2015F, 2016F 2017F.

  • M1522.001000 Computer Vision, 2016S, 2017S.

  • 4190.101 Discrete Mathematics, 2016S, 2017S.

  • 4190.429 Image Processing, 2015S.