Gunhee Kim, PhD (김건희)

Gunhee Kim at DC 

Associate 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

  • 인턴 및 대학원 지원자 상시 모집합니다: 자세한 사항은 먼저 이 문서를 읽기 바랍니다.

  • For foreign students: Unless you are already in Korea with valid Korean visa and can work more than 6 months, 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].

  • 본 연구실에서 spin-off 스타트업 rippleAI 에서 part-time 및 full-time 개발자 상시 채용합니다. [링크]


Selected Projects

Gunhee Kim, ICML2021 Unsupervised Representation Learning via Neural Activation Coding
  Yookoon Park (Columbia), Sangho Lee, Gunhee Kim and David Blei (Columbia)
  ICML 2021 (Long) [code] [pdf]
Gunhee Kim, ICML2021 Unsupervised Skill Discovery with Bottleneck Option Learning
  Jaekyeom Kim, Seohong Park and Gunhee Kim
  ICML 2021 [code] [pdf]
Gunhee Kim, NAACL2021 How Robust are Fact Checking Systems on Colloquial Claims?
  Byeongchang Kim*, Hyunwoo Kim*, Seokhee Hong and Gunhee Kim
  NAACL-HLT 2021 [code] [pdf]
Gunhee Kim, CVPR2021 Transitional Adaptation of Pretrained Models for Visual Storytelling
  Youngjae Yu*, Jiwan Chung*, Jongseok Kim, Heeseung Yun and Gunhee Kim
  CVPR 2021 [code] [pdf]
Gunhee Kim, CVPR2021 StyleMix: Separating Content and Style for Enhanced Data Augmentation
  Minui Hong*, Jinwoo Choi* and Gunhee Kim
  CVPR 2021 [code] [pdf]
Gunhee Kim, ICLR2021 SEDONA: Search for Decoupled Neural Networks toward Greedy Block-wise Learning
  Myeongjang Pyeon, Jihwan Moon, Taeyoung Hahn and Gunhee Kim
  ICLR 2021 [code] [pdf]
Gunhee Kim, ICLR2021 Drop-Bottleneck: Learning Discrete Compressed Representation for Noise-Robust Exploration
  Jaekyeom Kim, Minjung Kim, Dongyeon Woo and Gunhee Kim
  ICLR 2021 [code] [pdf]
Gunhee Kim, ICLR2021 Parameter Efficient Multimodal Transformers for Video Representation Learning
  Sangho Lee, Youngjae Yu, Gunhee Kim, Thomas Breuel (NVIDIA), Jan Kautz (NVIDIA) and Yale Song (Microsoft Research)
  ICLR 2021 [code] [pdf]
Gunhee Kim, ICLR2021 Self-Supervised Learning of Compressed Video Representations
  Youngjae Yu*, Sangho Lee*, Gunhee Kim and Yale Song (Microsoft Research)
  ICLR 2021 [pdf]
Gunhee Kim, AAAI2021 Dual Compositional Learning in Interactive Image Retrieval
  Jongseok Kim, Hoeseong Kim, Youngjae Yu and Gunhee Kim
  AAAI 2021 [code] [pdf]
Gunhee Kim, AAAI2021 IB-GAN: Disentangled Representation Learning with Information Bottleneck GAN
  Insu Jeon, Wonkwang Lee, Myeongjang Pyeon and Gunhee Kim
  AAAI 2021 [pdf]
Gunhee Kim, EMNLP2020 Will I Sound Like Me? Improving Persona Consistency in Dialogues through Pragmatic Self-Consciousness
  Hyunwoo Kim, Byeongchang Kim and Gunhee Kim
  EMNLP 2020 [code] [pdf]
Gunhee Kim, ECCV2020 Model-Agnostic Boundary-Adversarial Sampling for Test-Time Generalization in Few-Shot learning
  Jaekyeom Kim, Hyoungseok Harry Kim and Gunhee Kim
  ECCV 2020 (Oral) [code] [pdf]
Gunhee Kim, ECCV2020 Character Grounding and Re-Identification in Story of Videos and Text Descriptions
  Youngjae Yu, Jongseok Kim, Heeseung Yun, Jiwan Chung and Gunhee Kim
  ECCV 2020 (Spotlight) [code] [pdf]
Gunhee Kim, ECCV2020 Rethinking Class Activation Mapping for Weakly Supervised Object Localization
  Junhyug Noh*, Wonho Bae* and Gunhee Kim
  ECCV 2020 [code] [pdf]
Gunhee Kim, ECCV2020 Imbalanced Continual Learning with Partitioning Reservoir Sampling
  Chris Dongjoo Kim, Jinseo Jeong and Gunhee Kim
  ECCV 2020 [code] [pdf]
Gunhee Kim, ICLR2020a Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue
  Byeongchang Kim, Jaewoo Ahn and Gunhee Kim
  ICLR 2020 (Spotlight) [code] [pdf]
Gunhee Kim, ICLR2020b A Neural Dirichlet Process Mixture Model for Task-Free Continual Learning
  Soochan Lee, Junsoo Ha, Dongsu Zhang and Gunhee Kim
  ICLR 2020 [code] [pdf]
Gunhee Kim, NIPS2019 Self-Routing Capsule Networks
  Taeyoung Hahn, Myeongjang Pyeon and Gunhee Kim
  NeurIPS 2019 [code] [pdf]
Gunhee Kim, VRST2019 POL360: A Universal Mobile VR Motion Controller using Polarized Light
  Hyouk Jang, Juheon Choi and Gunhee Kim
  ACM VRST 2019 [pdf] [demo]
  (Best Paper Award - Runner-up)
Gunhee Kim, ICCV2019 Better to Follow, Follow to be Better: Towards Precise Supervision of Feature Super-Resolution for Small Object Detection
  Junhyug Noh, Wonho Bae, Wonhee Lee, Jinhwan Seo and Gunhee Kim
  ICCV 2019 [code] [pdf]
Gunhee Kim, ICML2019 Variational Laplace Autoencoders
  Yookoon Park, Chris Dongjoo Kim and Gunhee Kim
  ICML 2019 [code] [pdf]
Gunhee Kim, ICML2019 Curiosity-Bottleneck: Exploration By Distilling Task-Specific Novelty
  Youngjin Kim, Hyunwoo Kim*, Wontae Nam* (KC ML2), Ji-Hoon Kim (NAVER) and Gunhee Kim
  ICML 2019 [code] [pdf]
Gunhee Kim, CVPR2019 Multi-task Self-supervised Object Detection via Recycling of Bounding Box Annotations
  Wonhee Lee, Joonil Na and Gunhee Kim
  CVPR 2019 [code] [pdf]
Gunhee Kim, NAACL2019 AudioCaps: Generating Captions for Audio in the Wild
  Chris Dongjoo Kim, Byeongchang Kim, Hyunmin Lee and Gunhee Kim
  NAACL-HLT 2019 (Oral) [code] [pdf]
Gunhee Kim, NAACL2019 Abstractive Summarization of Reddit Posts with Multi-level Memory Networks
  Byeongchang Kim, Hyunwoo Kim and Gunhee Kim
  NAACL-HLT 2019 (Oral) [code] [pdf]
Gunhee Kim, ICLR2019 Harmonizing Maximum Likelihood with GANs for Multimodal Conditional Generation
  Soochan Lee, Junsoo Ha and Gunhee Kim
  ICLR 2019 [code] [pdf]
Gunhee Kim, ICLR2019 Discovery of Natural Language Concepts in Individual Units
  Seil Na, Yo Joong Choe (KaKao Brain), Dong-Hyun Lee (KaKao Brain), Gunhee Kim
  ICLR 2019 [code] [pdf]
Gunhee Kim, ECCV2018 A Joint Sequence Fusion Model for Video Question Answering and Retrieval
  Youngjae Yu, Jongseok Kim and Gunhee Kim
  ECCV 2018 [code] [pdf]
Gunhee Kim, ICML2018 Video Prediction with Appearance and Motion Conditions
  Yunseok Jang, Gunhee Kim, and Yale Song (Microsoft Research)
  ICML 2018 [code] [pdf]
Gunhee Kim, CVPR2018 Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors
  Junhyug Noh, Soochan Lee, Beomsu Kim, and Gunhee Kim
  CVPR 2018 [code] [pdf]
Gunhee Kim, CVPR2018 A Memory Network Approach for Story-based Temporal Summarization of 360° Videos
  Sangho Lee, Jinyoung Sung, Youngjae Yu, and Gunhee Kim
  CVPR 2018 [code] [pdf]
Gunhee Kim, NAACL2018 A Hierarchical Latent Structure for Variational Conversation Modeling
  Yookoon Park, Jaemin Cho and Gunhee Kim
  NAACL-HLT 2018 (Oral) [code] [pdf]
Gunhee Kim, ICLR2018 Memorization Precedes Generation: Learning Unsupervised GANs with Memory Networks
  Youngjin Kim, Minjung Kim, and Gunhee Kim
  ICLR 2018 [code] [pdf]
Gunhee Kim, AAAI2018 A Deep Ranking Model for Spatio-Temporal Highlight Detection from a 360° Video
  Youngjae Yu, Sangho Lee, Joonil Na, Jaeyun Kang, and Gunhee Kim
  AAAI 2018 [pdf]
Gunhee Kim, ICCV2017 A Read-Write Memory Network for Movie Story Understanding
  Seil Na, Sangho Lee, Jisung Kim (SKT), and Gunhee Kim
  ICCV 2017 [code] [pdf]
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 (KAIST) (*: equal contribution)
  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]
  A longer version is published in IEEE TPAMI vol.41, no.4, pp.999-1012, April 2019 [link]
  Towards Personalized Image Captioning via Multimodal Memory Networks
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]
  A longer version is accepted in IJCV vol.127, no.10, pp.1385-1412, October 2019 [link]
  Video Question Answering with Spatio-Temporal Reasoning
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 [pdf]
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) [code] [pdf]
  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 [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 is published in IEEE TPAMI vol.40, no.4, pp.945-957, April 2018 [link]
  Retrieval of Sentence Sequences for an Image Stream via Coherence Recurrent Convolutional Networks
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



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.