Gyeongmin Choe (최경민)


#211, ITC Building (N1), Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon, Korea                  
Email: gmchoe08@gmail.com   Phone: +82-10-9707-7517 


    I am a Ph.D candidate in KAIST majoring computer vision. My research interests include:  

    3D Computer Vision: Depth/Surface Normal Estimation, Depth Enhancement

    Depth Sensors: RGB-D Sensors (e.g.Kinect), Stereo Cameras, Light-field Cameras

    Multispectral ImagingImage Enhancement/Understanding using RGB-NIR Stereo

    Deep Learning: Deep Shape from Shading, Deep Semantic Segmentation



News



TPAMI and 1 RAL papers accepted                                                                                                                               2018

ICCV Doctoral Consortium Awarded & Joined in Venice, Italy                                                                                            2017

1 ICRA paper accepted                                                                                                                                                 2017

I won Best Poster Award (1st place) in IWRCV 2016  [Prize]                                                                                        2016

I won Grand prize (1st place) for BK 21 plus overseas study report competition  [Prize]                                                  2016

1 CVPR, 1 IJCV and 2 ECCV papers accepted                                                                                                                2016

Visiting Carnegie Mellon University to work with prof. Srinivasa Narasimhan                                         2015.07 - 2015.12

1 CVPR, 1 ICCV and 2 ICIP papers accepted                                                                                                                 2015

1 CVPR paper accepted                                                                                                                                                 2014

I Won 20th Samsung HumanTech Paper Award Silver prize (2nd place)  [Prize]                                                             2014

Educations



Ph.D., KAIST - Electrical Engineering                                                                                                        2013.02 - 2018.06

(Integrated master's and doctor's course)

Advisor: prof. In So Kweon 


Visiting Ph.D. student, Carnegie Mellon University - Robotics Institute                                                    2015.07 - 2015.12

Advisor: prof. Srinivasa Narasimhan


B.S., KAIST - Mechanical Engineering                                                                                                       2008.02 - 2012.02      
                    
High School Diploma (completed in two years), Kangwon Science High School                                             2006.03 - 2008.02


International Journals



 
Accurate 3D
 Reconstruction for Small Motion Clip for Rolling Shutter Cameras
,
Sunghoon Im, Hyowon Ha, Gyeongmin Choe, Hae-Gon Jeon, Kyungdon Joo and In So Kweon,
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2018

[project page] [pdf] [bibtex]

RANUS: RGB and NIR Urban Scene Dataset for Deep Scene Parsing,
Gyeongmin Choe, Seong-heum Kim, Sunghoon Im, Joon-Young Lee, Srinivasa Narasimhan, and In So Kweon 
IEEE Robotics and Automation Letters (RAL), Feb. 2018
[project page] [pdf] [bibtex]

Depth from a Light Field Image with Learning-based Matching Costs,

Hae-Gon Jeon, Jaesik Park, Gyeongmin Choe, Jinsun Park, Yunsu Bok, Yu-Wing Tai, and In So Kweon,  
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2018
[project page] [pdf] [bibtex]

Refining Geometry from Depth Sensors using IR Shading Images,
Gyeongmin Choe, Jaesik Park, Yu-Wing Tai and, In So Kweon,  
International Journal of Computer Vision (IJCV), Mar. 2017
[project page] [pdf] [bibtex]


International Conferences



 

A Taxonomy and Evaluation of Dense Light Field Depth Estimation Algorithms ,
Ole Johannsen, Katrin Honauer, Bastian Goldluecke, Anna Alperovich, Federica Battisti, Yunsu Bok, Michele Brizzi, Marco Carli, Gyeongmin Choe, Maximilian Diebold, Marcel Gutsche, Hae-Gon Jeon, In So Kweon, Jaesik Park, Jinsun Park, Hendrik Schilling, Hao Sheng, Lipeng Si, Michael Strecke, Antonin Sulc, Yu-Wing Tai, Qing Wang, Ting-Chun Wang, Sven Wanner, Zhang Xiong, Jingyi Yu, Shuo Zhang, and Hao Zhu,
IEEE CVPR Workshop - Light Field for Computer Vision, Jul. 2017.
[project page] [pdf] [bibtex]

A Real-Time and Energy-Efficient Embedded System for Intelligent ADAS with RNN-Based Deep Risk Prediction using Stereo Camera,
Kyuho Lee, Gyeongmin Choe, Kyeongryeol Bong, Changhyeon Kim, In So Kweon, and Hoi-Jun Yoo
International Conference on Computer Vision Systems (ICVS), Jul. 2017.
[project page] [pdf] [bibtex]

Deep Representation of Industrial Components using Simulated Images,
Seong-heum Kim, Gyeongmin Choe, Byungtae Ahn, and In So Kweon, 
IEEE International Conference on Robotics and Automation (ICRA), May. 2017.
[project page] [pdf] [bibtex]

Fine-scale Surface Normal Estimation using a Single NIR Image,
Youngjin Yoon*, Gyeongmin Choe*, Namil Kim, Joon-Young Lee, and In So Kweon, 
European Conference on Computer Vision (ECCV), Oct. 2016. 
* Equal contiribution
[project page] [pdf] [bibtex]
All-around Depth from Small Motion with A Spherical Panoramic Camera,
Sunghoon Im, Hyowon Ha, Francois Rameau, Hae-Gon Jeon, Gyeongmin Choe, and In So Kweon, 
European Conference on Computer Vision (ECCV), Oct. 2016. 
[project page] [pdf] [bibtex]

Simultaneous Estimation of Near IR BRDF and Fine-Scale Surface Geometry,
Gyeongmin Choe, Srinivasa G. Narasimhan, and In So Kweon, 
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016. Spotlight (Acceptance rate of 9.7%)
[project page] [pdf] [bibtex]
High Quality Structure from Small Motion for Rolling Shutter Cameras ,
Sunghoon Im, Hyowon Ha, Gyeongmin Choe, Hae-Gon Jeon, Kyungdon Joo, and In So Kweon ,
IEEE International Conference on Computer Vision (ICCV), Dec. 2015.
[project page] [pdf] [bibtex]

Depth from Accidental Motion using Geometry Prior ,
Sunghoon Im, Gyeongmin Choe, Hae-Gon Jeon, and In So Kweon, 
IEEE International Conference on Image Processing (ICIP) [Top 10% paper], Sep. 2015. 
[project page] [pdf] [bibtex]

Reflection Removal using Disparity and Gradient-Sparsity via Smoothing Algorithm ,
Tharatch Sirinukulwattana, Gyeongmin Choe, and In So Kweon, 
IEEE International Conference on Image Processing (ICIP), Sep. 2015 
[pdf] [bibtex]

Accurate Depth Map Estimation from a Lenslet Light Field Camera,
Hae-Gon Jeon, Jaesik Park, Gyeongmin Choe, Jinsun Park, Yunsu Bok, Yu-Wing Tai, and In So Kweon,
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2015.

[project page] [pdf] [bibtex]


Exploiting Shading Cues in Kinect IR Images for Geometry Refinement,
Gyeongmin Choe, Jaesik Park, Yu-Wing Tai, and In So Kweon, 
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2014.
[project page] [pdf] [bibtex] [codes]
 

Reviewer Experiences



- IEEE  International Conference on Computer Vision and Pattern Recognition (CVPR)

- IEEE  International Conference on Multimedia Information Processing and Retrieval (MIPR)

- Elsevier - Computer Vision and Image Understanding (CVIU)

- DEStech - Computer Science and Application Engineering (CSAE)

- Springer - Computational Visual Media Conference (CVM)


Awards & Honors



- Grant, ICCV Doctoral Consortium, Venice, Italy, Oct. 2017

Depth estimation challenge: robustness champion, CVPR workshop on Light Fields for Computer Vision, Jul. 2017 [Prize]

Best Poster Presentation Award, IPIU 2017, Korea, Feb. 2017 [Prize]

Best Poster Award, The 11th IWRCV, Korea, Dec. 2016 [Prize] 

- Grand prize (1st place), BK 21 plus overseas study report competition, KAIST, Jul. 2016 [Prize] 

- Best Poster Award, The 10th IWRCV, China, Nov. 2015 [Prize] 

- Official Best 10% Paper Selection, ICIP 2015 Organizing Committee, Sep. 2015 

- Oral presentation, The 9th IWRCV, Japan, Dec. 2014 

- Certificate, International Computer Vision Summer School (ICVSS 2014), Sicily, Italy, Jul. 2014 [Certificate] 

- Silver prize(2nd place), 20th HumanTech Paper Award, Samsung Electronics Corp., Feb. 2014 [Prize] 

- Student volunteer, Asian Conference on Computer Vision (ACCV 2012), Daejeon, Korea, Nov. 2012 

- Full scholarship, KAIST, Feb. 2008 ~ Feb. 2012 

- Scholarship, Samsung Electronics Corp., Feb. 2008 

- Silver Prize(2nd place), 13th HumanTech Paper Award, Samsung Electronics Corp., Feb. 2007


Patents



METHOD AND APPARATUS OF FINE-SCALE SURFACE NORMAL ESTIMATION USING SINGLE NIR IMAGE, AND NEURAL NETWORK LEARNING METHOD

kr 10-2017-0112107 / Application date: Sep. 2017 

APPARATUS AND METHOD FOR DEPTH MAP GENERATION
kr 10-1745493-0000 / Registration date: Jun. 2017
DEVICE AND METHOD FOR OBTAINING ACCURATE 3D INFORMATION USING DEPTH SENSOR AND INFRARED SHADING CUES
kr 10-1707939-0000 / Registration date: Feb. 2017
METHOD AND APPARATUS FOR ACQUIRING DEPTH MAP FROM ALL-AROUND CAMERA
kr 10-2016-0167525 / Application date: Dec. 2016 
METHOD AND SYSTEM FOR ESTIMATING SURFACE GEOMETRY AND REFLECTANCE
kr 10-2016-0167526 / Application date: Dec. 2016  

특허