Dong-Jin Kim (김동진)

Ph.D Candidate.

Robotics and Computer Vision Lab.

School of Electrical Engineering

Korea Advanced Institute of Science and Technology (KAIST)

Daejeon, South Korea.

Email : djnjusa [at] kaist.ac.kr / djnjusa [at] gmail.com

Office : Room 212, N1 ITC Building

Phone: (+82)-42-350-5465

[CV] [linkedin]

I'm a Ph.D. student advised by Prof. In So Kweon (from March 2017 to present) in School of Electrical Engineering (EE) of KAIST (Korea Advanced Institute of Science and Technology), South Korea. I received my B.S and M.S degrees in EE from KAIST in 2015 and 2017 respectively. I was a student internship with researcher Xiao Sun and Steve Lin in Visual Computing Group, the Microsoft Research Asia (MSRA) from June 2019 to November 2019. I received Silver Prize of Samsung Humantech awards and Qualcomm Innovation award as the 1st author.

Research Interests

  • Visual Recognition
  • Language and Vision
  • Human Behavior Understanding
  • Data Issues in Deep Learning

News

Aug. 2019. One paper is accepted in EMNLP 2019.

Jun. 2019. Research Intern in MSRA starts.

Mar. 2019. One paper is accepted in CVPR 2019.

Selected Publications

"Image Captioning with Very Scarce Supervised Data: Adversarial Semi-Supervised Learning Approach"

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh, and In So Kweon.

International Conference on Empirical Methods in Natural Language Processing (EMNLP), 2019.

(23.8% acceptance rate)

[PDF] [Project]

  • Also, presented at "Language&Vision " and "Visual Question Answering and Dialog " Workshops in conjunction with CVPR 2019, and "CLVL: 3rd Workshop on Closing the Loop Between Vision and Language" in conjunction with ICCV 2019.

"Dense Relational Captioning: Triple-Stream Networks for Relationship-Based Captioning"

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh, and In So Kweon.

IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2019.

(25.2% acceptance rate)

[PDF] [Project] [Dataset] [code]

  • Received Qualcomm Innovation Award 2019.
  • Also presented at "Language&Vision" and "Visual Question Answering and Dialog" Workshops in conjunction with CVPR 2019.

"Disjoint Multi-task Learning between Heterogeneous Human-centric Tasks"

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh, Youngjin Yoon and In So Kweon.

IEEE Winter Conference on Applications of Computer Vision (WACV), Mar 2018. (Oral)

[PDF]

Other Publications

  • "Disjoint Multi-task Learning between Heterogeneous Human-centric Tasks"

Dong-Jin Kim, Jinsoo Choi, Tae-Hyun Oh, Youngjin Yoon and In So Kweon.

International Workshop on Robust Computer Vision (IWRCV), 2019

  • "Sentence Learning on Deep Convolutional Networks for Image Caption Generation"

Dong-Jin Kim, Donggeun Yoo, Bonggeun Sim, and In So Kweon.

International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Aug 2016.

  • "Learning Deep Convolutional Representations with Image-Level Sentences for Image Caption Generation"

Dong-Jin Kim, Donggeun Yoo, Bonggeun Sim, and In So Kweon.

Workshop on Image Processing and Image Understanding (IPIU), Feb 2016.

  • "Relative Attributes with Deep Convolutional Neural Network"

Dong-Jin Kim, Donggeun Yoo, Sunghoon Im, Namil Kim, T. Sirinukulwattana and In So Kweon.

International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Oct 2015.

Honors & Awards

  • Silver Prize, 26th Samsung Humantech Paper Awards (Top 1.6%) [certificate]

"Detecting Human-Object Interactions with Action Co-occurrence Prior"

Samsung Electronics Co., Ltd.

"Dense Relational Image Captioning via Multi-task Triple-Stream Networks"

Qualcomm Inc.

  • Certificate

International Computer Vision Summer School (ICVSS 2018) [certificate]

Sicily, Italy