Hyeongseok Son

Ph.D. student at POSTECH

Email: sonhs [at] postech.ac.kr

[Google Scholar], [Github]

Education

  • Ph.D (Joint M.S & Ph.D.) in CSE, POSTECH (Mar. 2014 ~ Aug. 2021)

    • Computer Graphics Lab.

    • Advisor: Prof. Seungyong Lee

    • Dissertation: Deep Image and Video Deblurring via Task-Specific Network Design

    • Research interests: image and video processing using deep learning, deblurring, stereo matching

  • B.S. in CSE, POSTECH (Mar. 2010 ~ Feb. 2014)

  • Jeonnam Science High School (Mar. 2008 ~ Feb. 2010)

Publications

International Conference & Journal

  • Hyeongseok Son, Junyong Lee, Sunghyun Cho, Seungyong Lee, “Single Image Defocus Deblurring Using Kernel-Sharing Parallel Atrous Convolutions”, In Proc. ICCV, 2021. (Accepted) [arxiv][code]

  • Hyeongseok Son, Junyong Lee, Jonghyeop Lee, Sunghyun Cho, Seungyong Lee, "Recurrent Video Deblurring with Blur-Invariant Motion Estimation and Pixel Volumes", ACM Transactions on Graphics (presented at SIGGRAPH 2021), Vol. 40, No. 5, 2021. [link][code]

  • Junyong Lee, Hyeongseok Son, Jaesung Rim, Sunghyun Cho, Seungyong Lee, "Iterative Filter Adpative Network for Single Image Defocus Deblurring", In Proc. CVPR, 2021. [PDF][code]

  • Junyong Lee, Hyeongseok Son, Gunhee Lee, Jonghyeop Lee, Sunghyun Cho, Seungyong Lee, "Deep Color Transfer using Histogram Analogy", The Visual Computer (special issue on CGI 2020), Vol. 36, No. 10, 2020. [PDF][code]

  • Hyeongseok Son, Gunhee Lee, Sunghyun Cho, Seungyong Lee, "Naturalness-Preserving Image Tone Enhancement Using Generative Adversarial Networks", Computer Graphics Forum (special issue on Pacific Graphics 2019), Vol. 38, No. 7, 2019. (Best Paper Award) [PDF][code]

  • *Beomseok Kim, *Hyeongseok Son, Seong-Jin Park, Sunghyun Cho, Seungyong Lee, "Defocus and Motion Blur Detection with Deep Contextual Features", Computer Graphics Forum (special issue on Pacific Graphics 2018), Vol. 37, No. 7, 2018. (*equal contribution) [PDF][code]

  • Seong-Jin Park, Hyeongseok Son, Sunghyun Cho, Ki-Sang Hong, Seungyong Lee, "SRFeat: Single Image Super-Resolution with Feature Discrimination", In Proc. ECCV, 2018. [project][code]

  • Hyeongseok Son and Seungyong Lee, "Fast Non-blind Deconvolution via Regularized Residual Networks with Long/Short Skip-Connections", In Proc. IEEE ICCP, 2017. [project][link][code]

International Workshop

  • Junyong Lee, Hyeongseok Son, Jonghyeop Lee, Seungyong Lee, "Computational Photography Software using Deep Learning", ICCV (Demo), 2019.

  • Junho Jeon, Seong-Jin Park, Hyeongseok Son, Seungyong Lee, "Computational Photography Software using Deep Learning: Perceptual Image Super-Resolution and Depth Image Refinement", ECCV (Demo), 2018.

  • Hyeongseok Son and Seungyong Lee, "Intrinsic Image Decomposition using Deep Convolutional Network", CVPR SUNw: Scene Understanding Workshop (Poster), 2016. [PDF]

  • Hyeongseok Son, Jaehwan Lee, Seungyong Lee, "Object Extraction via Stereo Matching and GrabCut", ICCV 1st International Workshop on Recovering 6D Object Pose (Poster), 2015. [PDF]

In Progress

  • *Hyeongseok Son, *Junyong Lee, Sunghyun Cho, Seungyong Lee, “Progressive Video Deblurring with Motion Compensation Using Structure-Detail Separation”. (*equal contribution)(submitted)

Korean Domestic Conference & Journal

  • Jonghyeop Lee, Junyong Lee, Hyeongseok Son, Seungyong Lee, “다중 스케일 심층 영상 프라이어 기반의 파노라마 영상 인페인팅”, In Proc. IPIU, 2021.

  • Hyeongseok Son, Jonghyeop Lee, Seungyong Lee, “움직이는 물체에 강인한 영상 스티칭”, In Proc. IPIU, 2021.

  • Junyong Lee, Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Single Image Defocus Deblurring using Dynamic Filter Network”, In Proc. KSC, 2020. (우수발표논문상)

  • Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Perceptual Quality Enhancement for Degraded Images”, In Proc. KSC, 2020.

  • Jonghyeop Lee, Hyeongseok Son, Junyong Lee, Haeun Yoon, Sunghyun Cho, Seungyong Lee, “Single Panorama Depth Estimation using Domain Adaptation”, Journal of the Korea Computer Graphics Society, Vol.26, No.3, 2020.

  • Jaepil Yu, Hojin Cho, Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Text Image Super-Resolution using Text Recognition Information”, In Proc. IPIU, 2019. (우수포스터상)

  • Jaepil Yu, Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Image Super Resolution with Text Handling Via Generative Adversarial Network”, KIISE Transactions on Computing Practices, Vol.24, No.8, 2018.

  • Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Non-Uniform Blur Removal for Single Image using Encoder-Decoder Network”, In Proc. KCGS, 2018. (포스터 발표 최우수상)

  • Jaepil Yu, Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “Image Super Resolution with Text Handling Via Generative Adversarial Network”, In Proc. KSC, 2017. (우수발표논문상)

  • Hyeongseok Son, Sunghyun Cho, Seungyong Lee, “HDR Style Brightness Adjustment for a Single Image”, In Proc. KSC, 2017.

  • Jaehwan Lee, Hyeongseok Son, Seungyong Lee, “대립적 네트워크를 사용한 얼굴 보정”, In Proc. IPIU, 2017.

  • Hyeongseok Son, Seungyong Lee, “Non-blind Deconvolution via Deep Convolutional Network”, In Proc. KCGS, 2016.

  • Hyeongseok Son, Jaehwan Lee, Seungyong Lee, “영상 채색 기술을 이용한 스테레오 매칭”, In Proc. IPIU, 2016.

  • Jaehwan Lee, Hyeongseok Son, Seungyong Lee, “Enhancement of Stereo Matching by Object Edge Refinement, In Proc. KCC, 2015.

  • Hyeongseok Son, Seungyong Lee, “알파맵을 이용한 영상 에지 개선”, In Proc. IPIU, 2015. (포스터 발표 최우수상)

Patents

  • Seungyong Lee, Sunghyun Cho, and Hyeongseok Son, Junyong Lee, "METHOD AND APPARATUS FOR VIDEO RESTORATION BASED ON MACHINE LEARNING", KR 10-2020-0180046 (filed for a patent), 2020.12.

  • Seungyong Lee, Sunghyun Cho, and Hyeongseok Son, Junyong Lee, "PIXEL VOLUME-BASED MACHINE LEARNING METHOD FOR VIDEO QUALITY ENHANCEMENT AND APPARATUS USING SAME", KR 10-2020-0188668 (filed for a patent), 2020.12.

  • Seungyong Lee, Sunghyun Cho, and Hyeongseok Son, "METHOD FOR PROCESSING IMAGE BASED ON MACHINE LEARNING AND APPARATUS THEREFOF", PCT/KR2020/015722 (filed for a patent), 2020.11.

  • Seungyong Lee, Sunghyun Cho, and Hyeongseok Son, "METHOD FOR PROCESSING IMAGE BASED ON MACHINE LEARNING AND APPARATUS THEREFOF", KR 10-2019-0178401 (filed for a patent), 2019.12.

  • Seungyong Lee, Seong-Jin Park, Hyeongseok Son, "MACHINE LEARNING METHOD FOR RESTORING SUPER-RESOLUTION IMAGE", KR 10-2169242 (issued), 2020.10.

  • Seungyong Lee and Hyeongseok Son, "IMAGE PROCESSING APPARATUS AND METHOD USING MACHINE LEARNING", US 10,360,664 (issued), 2019.07.

  • Seungyong Lee and Hyeongseok Son, "APPARATUS AND METHOD FOR IMAGE PROCESSING", KR 10-1871098 (issued), 2018.06.

  • Seungyong Lee, Inhee Kim, Jaehwan Lee, Hyeongseok Son, "AN IMAGE PROCESSING DEVICES AND A COMPUTER READABLE RECORDING MEDIUM RECORDING AN IMAGE PROCESSING PROGRAM", KR 10-1813371 (issued), 2017.12.

Honors

  • Samsung Sponsorship Program at DS, 2020

  • Naver PhD Fellowship, 2019

  • Pacific Graphics 2019 Best Paper Award, 2019

Projects

  • Image Quality Assessment Using Deep Learning
    System LSI, Samsung Electronics Co., South Korea
    (Jan 2017 - Aug 2020)

  • Depth Estimation Using Stereo Camera and Its Applications
    DM Division, Samsung Electro-Mechanics Co., South Korea
    (May 2015 - Oct 2015)

  • Software Development for Big Visual Data Based High Quality Photograph Makeup
    SW Star Lab, Institute for Information & communications Technology Promotion (IITP), South Korea
    (Mar 2015 -
    Current)

  • Image Quality Enhancement via Big Image Data Analysis
    DMC R&D Center, Samsung Electronics Co., South Korea
    (July 2014 - Dec 2014)

Teaching Experience

  • Teaching assistance for Programming Languages (CSED321, POSTECH), Mar-Jun 2015

  • Teaching assistance for Computer Graphics (CSED451, POSTECH), Mar-Jun 2016